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

 

JUN z 022%10;

[010

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2/05 p:/CIRC/DaIeDue.indd-p.1

STUDIES IN REACT IVE EXTRUSION PROCESSING OF BIODEGRADABLE
POLYNIERIC MATERIALS

By

Sunder Balakrishnan

A DISSERTATION

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

DOCTOR OF PHILOSOPHY
Department of Chemical Engineering

2004

ABSTRACT

STUDIES IN REACT IVE EXT RUSION PROCESSING OF BIODEGRADABLE
POLYMERIC MATERIALS

By
Sunder Balakrishnan

Various reaction chemistries such as Polymerization, Polymer cross-linking and Reactive
grafting were investigated in twin-screw extruders. Poly (1,4—dioxan-2-one) (PPDX) was
manufactured in melt by the continuous polymerization of 1,4-dioxan-2-one (PDX)
monomer in a twin-screw extruder using Aluminum tri-sec butoxide (ATSB) initiator.
Good and accurate control over molecular weight was obtained by controlling the ratio of
monomer to initiator. A screw conﬁguration consisting of only conveying elements was
used for the polymerization. The polymerization reaction was characterized by a
monomer-polymer dynamic equilibrium, above the melting temperature of the polymer,
limiting the equilibrium conversion to 78-percent. Near complete (~100-percent)
conversion was obtained on co-polymerizing PDX monomer with a few moi-percent
(around 8-percent) Caprolactone (CL) monomer in a twin-screw extruder using ATSB
initiator. The co-polymers exhibited improved thermal stability with reduction in glass
transition temperature. The extruder was modeled as an Axial Dispersed Plug Flow
Reactor for the polymerization of CL monomer using Residence Time Distribution
(RTD) Analysis. The model provided a good ﬁt to the experimental RTD and conversion
data.

Aliphatic and aliphatic-aromatic co-polyesters, namely Polycaprolactone (PCL)
and Poly butylenes (adipate—co-terephthalate) (Ecoﬂex) were cross-linked in a twin-screw

extruder using radical initiator to form micro-gel reinforced biodegradable polyesters.

 

Cross-linked Ecoflex was further extrusion blended with talc to form blends suitable to
be blown into ﬁlms. A screw conﬁguration consisting of conveying and kneading
elements was found to be effective in dispersion of the talc particles (5-10 microns) in the
polyester matrix. While the rates of crystallization increased for the talc ﬁlled polyester
blends, overall crystallinity reduced. Mechanical, tear and puncture properties of ﬁlms
made using the talc ﬁlled polyester blends were comparable to linear low density
Polyethylene (LDPE).

Ecoflex-Thermoplastic Starch (TPS) graft co-polymers were continuously
manufactured in melt with maleic acid catalyst using a twin-screw co-rotating extruder.
The graft co-polymer was completely extractable in Dichloromethane and formed
transparent ﬁlms on solvent casting. Regular com-starch was maleated in a twin-screw
extruder using maleic anhydride or maleic acid, glycerol plasticizer and optional radical
initiator. Conﬁrmation of reactivity of maleic acid onto the starch backbone was
conﬁrmed by Fourier Transformed Infra Red (FTIR) Spectroscopy. The Maleated
Thermoplastic Starch (MTPS) samples were easier to process than TPS and exhibited
lower intrinsic viscosity values. Graft co-polymers of Ecoflex and cross-linked Ecoflex
with MTPS were continuously manufactured in a twin-screw extruder. Mechanical, tear
and puncture properties of ﬁlms made using the graft co-polymers were comparable to
LDPE. All the above-mentioned reactive extrusion technologies used in the synthesis of

novel biodegradable polymeric materials ﬁnd use in single-use plastics applications.

Sri Seshadri Swamigal Thunai

Dedicated to

My Parents

iv

ACKNOWLEDGEMENTS
First and foremost, I would like to thank Dr.Ramani Narayan; my advisor, guide and
mentor. His encouragement, motivation, advice and guidance have been astounding and I
am very fortunate to work under Dr.Narayan. Apart from my research, Dr. Narayan and
Mrs. Homa Narayan have been a great moral support for me during the course of my
Ph.D. I would also like to thank my Ph.D. committee; Dr. Kris Berglund, Dr. Dennis
Miller, Dr. Gregory Baker and Dr. Dan Graiver for their advice and support during the
course of my Ph.D. A special word of thanks should go to Ken Farminer, for his valuable
inputs and encouragement. I wish to acknowledge the support by Dr.Philippe Dubois
towards my Ph.D. My lab mates; Phuong Tran, Dr.Raquez, Yogaraj Nabar, Chisa
Brookes, Laura Fisher, Guoren Cheng, Madhu Srinivasan, deserve special mention. I
have enjoyed working with all these people who provided a wonderful atmosphere at
SINAS lab. I extend my heartfelt gratitude to all the people who collaborated with me
during the course of my Ph.D.; Dr.Jean-Marie Raquez, Dr.Boo-Young Shin and
Chandrika Rajagopalan. My gratitude extends to all these people who have helped me
with various aspects of my research; Chris Schuster — Extrusion; Christina Berger -
characterization and testing; Mario -Extrusion experiments; Tianqi Li - GPC
experiments; Brian Rook -extrusion experiments involving the side feeder; Rob Selden —
Mechanical issues concerning extruders; Krithika- Film Processing and UV-
Spectroscopy; Dr.Jayaraman, Bhaskar Patham and Darrell Marchant- Capillary
Rheometer Studies; Dr.Mackay and Anish- Dynamic Light Scattering experiments;
Shirish Karande and Khayyam Ali- Thesis formatting in Microsoft Word and Mohar—

Mathematical modeling.

My sincere thanks to Nancy Albright, who has been a great source of help to me.
Nancy has answers to all my questions and I seem to forget all my problems when I talk
to her. My gratitude to all the working staff of “DECS”- Division of Engineering
Computing Services for having provided me with additional hard drive space to
accommodate my research work. I would like to also thank Sachin Vaidya, Lavanya
Parthasarthy and Bhaskar Patham with whom I have had innumerable conversations and
discussions pertaining to various subjects in Chemical Engineering. I would like to
extend my heartfelt thanks to all my friends who have provided me with support and help
during the last ﬁve years- Anand, Ganesh Kumar, Swaminathan, Kamlesh Shah, Kannan
Balasubramanian, Seetha Balasubramanian, Skanth Ganesan, Sudarsanam Chellappa,
Jayakrishnan Swaminathan, Rajeev Kalyanaraman, Sreevatsan Kidambi, Chirag Shah,
Shirish Karande, Yash Parulekar, Aishwarya Parulekar, Kantha Kumar Pongaliur,
Akshay Kamath, Prassanna Balasundaram, Bhavana Bhoovaraghan, Gayathri Santhanam,
Venkata Giri Kolli, Prema Sonthalia and Kiran Misra. A special thanks to my best friend,
Srinivasan Rakhunathan, for being there for me whenever I needed him. He is a
wonderful person and I will cherish his friendship for the rest of my life.
This work would not have been possible without the help and support from my
“little family”. I can strongly say that this work was made possible by my parent’s
constant prayers, support, motivation, encouragement and belief. I would also like to
thank my sister Mrs. Vanitha Kumar for her constant source of encouragement. I am
grateful to God for giving me such a loving sister. A special thanks to my bro—in-law Mr.
R.S. Kumar; nephews Nirek and Medheer, my uncle Mr. Balasundaram, late Mr. Rajan,

Mr. Rajagopalan and Mrs. Susheela Rajagopalan, who have been a constant source of

vi

motivation for me. Last but not the least; I would like to thank my ﬁancee, Vijaya
Rajagopalan, who has been instrumental in my entire thesis. She has been a good listener,
great encourager and provided tremendous support during my Ph.D. work. Whether it
was blowing ﬁlm at Packaging Building or writing my thesis documentation, she was
with me throughout and this work would not have been possible without her. Although I
have mentioned only a few names, there are many more people who have helped me in
this venture and I would like to extend my appreciation to all the other people who I may

have left out.

vii

TABLE OF CONTENTS

Chapter 1 ............................................................................................................................. 1
1.1 RATIONALE ...................................................................................................... 1
1.2 PROPOSED GOALS .......................................................................................... 4
1.3 SPECIFIC OBJECTIVES ................................................................................... 5
1.4 STRUCTURE OF THESIS ................................................................................. 7

Chapter 2 ............................................................................................................................. 9

BULK EXTRUSION POLYMERIZATION OF 1, 4—DIOXAN-2-ONE .......................... 9
2.1 INTRODUCTION .............................................................................................. 9

2.1.1 Mechanism of Polymerization of PDX monomer .................................... 14
2.1.2 Lactone Polymerization Catalysts ............................................................. 20
2.1.3 Kinetics of PDX Polymerization .............................................................. 22
2.1.4 Thermodynamics of Polymerization ......................................................... 27
2.1.5 Molecular Weight and Molecular Weight Distribution ............................ 30
2.1.6 Extrusion Studies ...................................................................................... 33
2.1.6.1 Reactive Extrusion Polymerization ....................................................... 36
2.1.6.2 Screw Conﬁguration ............................................................................. 37
2.1.6.3 Residence Time Distribution ................................................................ 40
2.1.6.3.] Extrusion Modeling ........................................................................ 44

2.2 EXPERIMENTAL DETAILS .......................................................................... 49
2.2.1 Materials and Equipment .......................................................................... 50
2.2.2 PDX monomer puriﬁcation ....................................................................... 50

viii

2.2.3 Batch Polymerization Studies ................................................................... 53
2.2.4 Determination of Extent of Monomer Conversion ................................... 56
2.2.5 Extrusion Polymerization Studies ............................................................. 57
2.2.5.1 Screw Conﬁguration ............................................................................. 63
2.2.5.2 Residence Time Distribution Studies .................................................... 65
2.2.6 Intrinsic Viscosity ..................................................................................... 66
2.2.7 Thermal Analysis ...................................................................................... 69
2.2.8 UV-Spectroscopy ...................................................................................... 70
2.3 RESULTS AND DISCUSSION ....................................................................... 71
2.3.1 Batch Polymerization Studies ................................................................... 71

2.3.2 Effect of extrusion process parameters on CL and PDX monomer

conversion ................................................................................................................. 72
2.3.3 Effect of CL addition on PDX Polymerization ......................................... 74
2.3.4 Intrinsic Viscosity Results ........................................................................ 76
2.3.5 Thermal Analysis Results ......................................................................... 78
2.3.6 Residence Time Distribution Studies ........................................................ 81
2.3.6.1 Reactive Extrusion Modeling ............................................................... 83
Chapter 3 ........................................................................................................................... 92
3.1 INTRODUCTION ............................................................................................ 92
3.1.1 Cross-linking Chemistry ........................................... .. ............................... 92
3.1.2 Cross-linking Principles ............................................................................ 94
3.1.3 Filled Composites ..................................................................................... 97
3.2 EXPERIMENTAL DETAILS ........................................................................ 100

ix

3.2.1 Materials and Equipment ........................................................................ 100

3.2.2 Extrusion Cross-linking Studies ............................................................. 104
3.2.3 Determination of gel fraction .................................................................. 106
3.2.4 Extrusion blending of Ecoﬂex with talc ................................................. 107
3.2.5 Thermal Analysis .................................................................................... 110
3.2.6 Environmental Scanning Electron Microscopy (ESEM) Studies ........... 110
3.2.7 Blown Film Processing ........................................................................... 111
3.2.8 Mechanical Property Determination ....................................................... 112
3.3 RESULTS AND DISCUSSION ..................................................................... 113
3.3.1 Cross-linking Studies .............................................................................. 113
3.3.1.1 Cross-linking Cherrristry ..................................................................... 113

3.3.1.2 Effect of Residence Time on gel fraction of cross-linked polyesters. 115

3.3.1.3 Effect of Initiator Concentration on the gel fraction of cross-linked

polyesters ............................................................................................................ 1 16
3.3.1.4 Thermal Analysis ................................................................................ 117
3.3.1.5 Environmental Scanning Electron Microscopy (ESEM) Results ....... 120
3.3.2 Polyester-talc blending studies ............................................................... 122

3.3.2.1 Effect of screw conﬁguration on extrusion processing of Ecoﬂex with

talc .......................................................................................... 122
3.3.2.2 Thermal Analysis Results ................................................................... 123
3.3.2.3 Environmental Scanning Electron Microscopy (ESEM) Results ....... 128
3.3.2.4 Mechanical Property Results .............................................................. 130

Chapter 4 ......................................................................................................................... 134

4. 1 BACKGROUND ............................................................................................ 134
4.1.1 Starch Theory .......................................................................................... 136
4.1.2 Starch blends with polyesters .................................................................. 143

4.1.2.1 Incompatible Blends ........................................................................... 145
4.1.2.2 Compatibilization Theory ................................................................... 148
4.1.3 Starch-Polyester Reactive Blends ........................................................... 150
4.1.4 Starch Derivatives ................................................................................... 151

4.2 EXPERIMENTAL .......................................................................................... 155
4.2.1 Materials and Equipment ........................................................................ 155
4.2.2 Ecoﬂex-Thermoplastic Starch graft copolymers .................................... 156

4.2.2.1 Synthesis of Ecoﬂex-TPS graft copolymers by reactive extrusion

processing ........................................................................................................... 156
4.2.2.2 Soxhlet Extraction Studies .................................................................. 159
4.2.3 Ecoﬂex-Maleated thermoplastic starch graft copolymers ...................... 159
4.2.3.1 Maleation of starch by reactive extrusion processing ......................... 159

4.2.3.2 Synthesis of Ecoﬂex-MTPS graft copolymers by reactive extrusion

processing ........................................................................................................... 163
4.2.3.3 Soxhlet Extraction Studies .................................................................. 165
4.2.3.4 Fourier Transformed Infrared Spectroscopy (FI‘IR) Studies .............. 165
4.2.3.5 Thermal Analysis Studies ................................................................... 166
4.2.3.6 Intrinsic Viscosity Studies .................................................................. 166
4.2.3.7 Blown Film Studies ............................................................................. 167

xi

4.2.3.8 Mechanical Property Determination ................................................... 167

RESULTS AND DISCUSSION ..................................................................... 168

4.3.1 Ecoﬂex-TPS graft copolymers ................................................................ 169

4.3.1.1 Trans-esteriﬁcation Reaction Chemistry in the Synthesis of Ecoflex-

TPS graft copolymers ......................................................................................... 169
4.3.1.2 Proof of Grafting by Soxhlet Extraction ............................................. 170
4.3.1.3 Proof of Grafting by FTIR Analysis ................................................... 171
4.3.1.4 Proof of grafting by Thermal Gravimetric (TGA) Analysis ............... 173
4.3.1.5 Mechanical Properties Results ............................................................ 174
4.3.2 Maleated Thermoplastic Starch (MTPS) ................................................ 177
4.3.2.1 Starch Maleation chemistry ................................................................ 177
4.3.2.2 Results on reactive extrusion processing of MTPS ............................ 178

4.3.2.3 Proof of reaction of maleic anhydride and maleic acid onto starch by

FTIR Analysis ..................................................................................................... 179
4.3.2.4 Thermal Analysis Results ................................................................... 181
4.3.2.5 Intrinsic viscosity Results ................................................................... 185
4.3.3 Ecoﬂex-MTPS graft copolymers ............................................................ 188

4.3.3.1 Trans-esteriﬁcation Reaction Chemistry in the synthesis of Ecoﬂex-

MTPS graft copolymers ...................................................................................... 188
4.3.3.2 Proof of grafting by Soxlaiet Extraction .............................................. 189
4.3.3.3 Proof of grafting by FT IR Analysis .................................................... 192
4.3.3.4 Proof of grafting by Thermal Analysis ............................................... 193

4.3.3.5 Blown Film Processing Results of Ecoflex-MTPS graft copolymers 195

xii

 

4.3.3.6 Mechanical Property Results of Ecoflex-MTPS graft copolymers ..... 197
Chapter 5 ......................................................................................................................... 206
5.1 SUMMARY .................................................................................................... 206
5.1.1 Extrusion polymerization of 1, 4-dioxan—2-one (PDX) monomer and co-
polymerization with e-caprolactone (CL) monomer ............................................... 207

5.1.2 Reactive extrusion cross-linking of biodegradable polyesters (Ecoﬂex and

PCL) and blending with talc. .................................................................................. 209
5.1.3 Ecoflex-Starch graft copolymers by reactive extrusion processing ........ 210
5.2 CONCLUSIONS ............................................................................................. 212

5.2.1 Extrusion polymerization of 1, 4-dioxan—2-one (PDX) monomer and co-
polymerization with e-caprolactone (CL) monomer ............................................... 212

5.2.2 Reactive extrusion cross-linking of biodegradable polyesters (Ecoﬂex and

PCL) and blending with talc. .................................................................................. 215
5.2.3 Ecoﬂex-Starch graft copolymers by reactive extrusion processing ........ 217
Chapter 6 ......................................................................................................................... 222
6.1 RECOMMENDATIONS ................................................................................ 222
6.1.1 Bulk Extrusion Polymerization of 1, 4-dioxan-2-one monomer ............ 222

6.1.2 Reactive Extrusion Cross-linking of Biodegradable Polyester and
Blending with Inorganic Fillers. ............................................................................. 223

6.1.3 Polyester-Starch graft copolymers by Reactive Extrusion Processing 224

xiii

 

LIST OF FIGURES

Figure 2-1: Oxidative cyclization of DEG to PDX and competing dehydration reactions

................................................................................................................................... 10
Figure 2-2: PDX monomer ............................................................................................... 11
Figure 2-3: Reactivity of PDX monomer .......................................................................... 13
Figure 2—4: Chain Propagation in Lactone Polymerization .............................................. 15
Figure 2-5: Cationic ROP Reaction Mechanism .............................................................. 16
Figure 2-6: Anionic ROP Reaction Mechanism ............................................................... 17
Figure 2-7: Coordination-Insertion Reaction Mechanism ................................................ 17
Figure 2-8: Coordination-Insertion Mechanism in the formation of PPDX polymer ....... 18
Figure 2-9: Intermolecular and Intramolecular trans-esteriﬁcation of PPDX .................. 20

Figure 2—10: Effect of Polymerization Temperature on the monomer conversion reached

at equilibrium for a monomer to initiator ratio of 600 and a polymerization time of

14 h ............................................................................................................................ 25
Figure 2-11: Dispersive and Distributive Mixing schematic ............................................ 39
Figure 2-12: Conventional screw conﬁguration ............................................................... 40
Figure 2-13: Schematic of Distillation setup .................................................................... 51
Figure 2-14: Schematic of Drying setup ........................................................................... 53
Figure 2-15: Schematic of Batch polymerization setup .................................................... 54

Figure 2-16: Three-arm structure of the PPDX polymer produced by reactive extrusion
polymerization of PDX monomer ............................................................................. 57

Figure 2-17: Process schematic for REX-PPDX production ............................................ 59

xiv

Figure 2-18: PDX monomer pump calibration curve (PDX monomer maintained between

32-35%) .................................................................................................................... 60
Figure 2-19: ATSB initiator solution calibration curve .................................................... 60
Figure 2-20: Screw Conﬁguration evaluated for the Synthesis of REX-PPDX ............... 64
Figure 2-21: UV Set Up Block Diagram .......................................................................... 70

Figure 2-22: DSC therrnogram of PPDX homopolymer and PCL—co-PPDX copolymers
obtained by reactive extrusion with different ﬁnal molar fractions in CL: FCL= 0.00
(a), 0.11 (b), 0.16 (c) and 0.20 (d) (heating rate = 10°C/min, under N2, only the
second run is considered) .......................................................................................... 78

Figure 2—23: Time-dependence of weight loss at 180°C for PPDX and PCL-co-PPDX
polymers obtained by reactive extrusion with different ﬁnal molar fractions in CL:
FCL= 0.00 (a), 0.11 (b), 0.16 (c) and 0.20 (d) (T = 180°C under a He flow) ............ 80

Figure 2-24: Rate of weight loss for PPDX and PCL-co—PPDX polymers obtained by

reactive extrusion with different ﬁnal molar fractions in CL: FCL= 0.00 (a), 0.11 (b),

0.16 (c) and 0.20 (d) (T = 180°C under a He flow) .................................................. 81
Figure 2-25: RTD curves at 130°C ................................................................................... 82
Figure 2-26: RTD curves at 170°C .................................................................................. 83

Figure 2—27: Axial Dispersed PFR model applied to RTD data for CL polymerization at
130°C (Screw speed of 125 RPM and monomer feed rate of 14 mllmin) ................ 85

Figure 2-28: Axial Dispersed PFR model applied to RTD data for CL polymerization at
130°C (Screw speed of 125 RPM and monomer feed rate of 22 ml/min) ................ 85

Figure 2-29: Axial Dispersed PFR model applied to RTD data for CL polymerization at

130°C (Screw speed of 225 RPM and monomer feed rate of 14 ml/min) ................ 86

XV

Figure 2-30: Axial Dispersed PFR model applied to RTD data for CL polymerization at
130°C (Screw speed of 225 RPM and monomer feed rate of 22 ml/min) ................ 86
Figure 2-31: Axial Dispersed PFR model applied to RTD data for CL polymerization at
170°C (Screw speed of 125 RPM and monomer feed rate of 14 ml/min) ................ 87
Figure 2-32: Axial Dispersed PFR model applied to RTD data for CL polymerization at
170°C (Screw speed of 125 RPM and monomer feed rate of 22 mllmin) ................ 87
Figure 2-33: Axial Dispersed PFR model applied to RTD data for CL polymerization at
170°C (Screw speed of 225 RPM and monomer feed rate of 14 mllmin) ................ 88
Figure 2-34: Axial Dispersed PFR model applied to RTD data for CL polymerization at
170°C (Screw speed of 225 RPM and monomer feed rate of 22 mllmin) ................ 88

Figure 2-35: Model vs. Experimental conversions in the polymerization of CL monomer

to form three-arm PCL polymer (Table 2-8). ........................................................... 91
Figure 3-1: Schematic showing cross-linked and uncross-linked networks ..................... 93
Figure 3-2: Chemical structure of PCL polyester ........................................................... 101
Figure 3-3: Chemical structure of aliphatic-aromatic copolyester ................................. 103

Figure 34: Theoretical percent Carbon dioxide evolution of Ecoﬂex under EN 13432 test

protocol. .................................................................................................................. 103
Figure 3-5: Screw Conﬁguration used for Cross-linking studies ................................... 106
Figure 3-6: Process Schematic of extrusion processing of Ecoflex with talc ................. 108
Figure 3-7: Blown Film Processing Schematic .............................................................. 112
Figure 3-8: Cross-linking chemistry by free radical reaction ......................................... 114

Figure 3-9: DSC results of Ecoﬂex and cross-linked Ecoﬂex (0.1-wt% Luperox 101). 117

xvi

Figure 3-10: TGA Results of Ecoﬂex and LEcoﬂex (cross-linked using 0.1-wt% Luperox

101) ......................................................................................................................... 1 19
Figure 3-11: ESEM Results of Ecoflex .......................................................................... 120
Figure 3-12: ESEM Results of LEcoﬂex ........................................................................ 121

Figure 3-13: Schematic of mixing provided by kneading blocks used for the extrusion
processing of Ecoﬂex with talc ............................................................................... 122
Figure 3—14: DSC Results of Ecoﬂexltalc (70/30; w/w) ................................................. 124
Figure 3-15: DSC Results of LEcoﬂex/talc (70/30; w/w) [LEcoﬂex made with 0.1-wt%
Luperox 101] ........................................................................................................... 125
Figure 3-16: T GA Results of Ecoﬂexltalc (70/30; w/w) ................................................ 126

Figure 3-17: TGA Results of LEcoflex/talc (70/30; w/w) [LEcoﬂex made with 0.1-wt%

Luperox 101] ........................................................................................................... 127
Figure 3-18: ESEM Results of Ecoﬂexltalc (70/30; w/w) .............................................. 128
Figure 3-19: ESEM Results of LEcoﬂex/talc (70/30; w/w) ........................................... 129
Figure 3-20: Tensile strength of Ecoﬂex and LEcoflex blends with talc ....................... 130
Figure 3-21: Modulus of Elasticity of Ecoﬂex and LEcoflex blends with talc .............. 131
Figure 3-22: Break Elongation of Ecoﬂex and LEcoﬂex blends with talc ..................... 132
Figure 4—1: Structure of amylose .................................................................................... 136
Figure 42: Structure of amylopectin .............................................................................. 137
Figure 4-3: Swelling of Starch granules ......................................................................... 139
Figure 4-4: ESEM image of starch ................................................................................. 141
Figure 4-5: ESEM image of thermoplastic starch ........................................................... 141
Figure 4-6: X-Ray diffractogram results of starch and thermoplastic starch .................. 142

xvii

Figure 4-7: Effect of composition and viscosity on phase morphology ......................... 147
Figure 4-8: Starch Reaction with half-esters of dicarboxylic acids ................................ 153
Figure 4-9: Screw Conﬁguration used for the synthesis of Ecoﬂex-TPS graft copolymers
................................................................................................................................. 157
Figure 4-10: Calibration curve for Glycerol pump used in the reactive maleation of starch
................................................................................................................................. 161

Figure 4-11: Acid-catalyzed trans-esteriﬁcation chemistry in the synthesis of Ecoﬂex-

TPS graft copolymers. ............................................................................................ 169
Figure 4-12: FTIR Spectrum of Ecoflex-TPS graft co-polymers ................................... 172
Figure 4-13: TGA Results of Ecoflex-TPS graft copolymers ........................................ 173
Figure 4-14: Tensile strength results of Ecoflex-TPS graft copolymers ........................ 174
Figure 4-15: Modulus of elasticity results of Ecoﬂex-TPS graft copolymers ................ 175
Figure 4-16: Break elongation results of Ecoflex-TPS graft copolymers ....................... 176
Figure 4-17: Maleation of starch using maleic anhydride .............................................. 178
Figure 4-18: FTIR results of MTPS samples using maleic acid modiﬁer ...................... 180
Figure 4-19: FI‘IR results of MTPS samples using maleic anhydride modiﬁer ............. 180
Figure 4-20: DSC Results of MTPS samples ................................................................. 182
Figure 4-21: TGA Results of MTPS samples ................................................................. 183

Figure 4-22: TGA Results of MTPS samples modiﬁed with 2.5%, 5% and 10% maleic

acid .......................................................................................................................... 184
Figure 4-23: Intrinsic viscosity of regular corn starch .................................................... 186
Figure 4-24: Intrinsic viscosity of TPS ........................................................................... 186

Figure 4-25: Intrinsic viscosity of MTPS (2.5% Maleic anhydride) .............................. 187

xviii

Figure 4-26: Intrinsic viscosity of MTPS (2.5% Maleic anhydride and 0.1% Luperox 101)
................................................................................................................................. 187

Figure 4-27: Trans-esteriﬁcation reaction chemistry in the synthesis of Ecoﬂex-MTPS

graft copolymers. .................................................................................................... 189
Figure 4-28: Fl‘ IR Results of Ecoﬂex-MTPS graft copolymers .................................... 192
Figure 4-29: DSC Results of Ecoflex-MTPS graft copolymer ....................................... 193
Figure 4-30: TGA Results of Ecoﬂex-TPS graft copolymer .......................................... 194
Figure 4-31: Blown ﬁlm pressures of Ecoflex-MTPS graft copolymers ........................ 196
Figure 4-32: Tensile Strength Results of Ecoflex-MTPS graft copolymers ................... 197
Figure 4-33: Modulus of Elasticity Results of Ecoflex-MTPS graft copolymers .......... 198
Figure 4-34: Break Elongation Results of Ecoflex-MTPS graft copolymers ................. 199

Figure 4-35: Tensile Strength Results of Ecoflex-MTPS (70/30) graft copolymers 2.5-
wt%, 5-wt% and 10-wt% maleic acid in MTPS ..................................................... 201

Figure 4-36: Modulus of Elasticity Results of Ecoﬂex-MTPS (70/30) graft copolymers
2.5-wt%, 5-wt% and 10-wt% maleic acid in MTPS ............................................... 202

Figure 4-37: Break Elongation Results of Ecoflex-MTPS (70/30) graft copolymers 2.5-

wt%, 5—wt% and 10-wt% maleic acid in MTPS ..................................................... 203

xix

 

LIST OF TABLES

Table 2-1: Properties of PDX monomer ........................................................................... 10
Table 2-2: Patented PDX end uses .................................................................................... 12

Table 2-3: Influence of the nature of the initiator on the yield of PPDX obtained in bulk

at 100°C ..................................................................................................................... 24
Table 2-4: Reactions performed in co—rotating twin screw extruders ............................... 35
Table 2-5: Extrusion Processing Conditions for CL polymerization at 150°C ................. 58

Table 2-6: Polymerization of PDX monomer by Reactive Extrusion: Determination of
Optimal Conditions. .................................................................................................. 61
Table 2-7: PCL-co-PDX by reactive extrusion at 130°C: Effect of the addition of limited
amounts of CL monomer .......................................................................................... 63
Table 2-8: Experimental Scheme to synthesize REX-PCL for RTD studies .................... 66
Table 2-9: Conversions of PDX and CL monomers in experimental conditions close to
Reactive Extrusion (Polymerization time = 2 minutes) ............................................ 72
Table 2-10: Polymerization of PDX monomer by Reactive Extrusion: Determination of
optimal Conditions .................................................................................................... 73
Table 2-11: Polymerization of PDX by reactive extrusion: Effect of the addition of
limited amounts of CL monomer .............................................................................. 75
Table 2-12: Intrinsic viscosity of PPDX and PCL-cu-PPDX polymers produced by

reactive extrusion (T =130°C and Screw Speed = 130 RPM) ................................... 77

XX

Table 2-13: DSC data of PPDX and PCL-co-PPDX polymers as obtained by reactive
extrusion (under nitrogen flow, from —100°C to 140°C at a heating rate of 10°C/min,
2“d scan) .................................................................................................................... 79

Table 2-14: Fit coefﬁcients ‘b’ predicting goodness of ﬁt of the Axial Dispersed PFR
model to experimental data in the case of CL polymerization ................................. 84

Table 2-15: Peclet Number values under different experimental processing conditions

(Table 2-8) in the synthesis of three-arm PCL polymer ........................................... 89
Table 3-1: Processing Conditions for Cross-linking PCL and Ecoﬂex .......................... 105
Table 3-2: Extrusion Processing Conditions for Ecoﬂex blending with talc .................. 109
Table 3-3: Blown Film Processing Conditions for talc ﬁlled polyester blends .............. 111

Table 34: Gel fraction results of LPCL and LEcoflex samples obtained from the vent and
the die. ..................................................................................................................... 115

Table 3-5: Gel fraction results of LEcoﬂex containing varying amounts of Luperox 101

................................................................................................................................. 1 16
Table 3—6: Tear and Puncture properties of Ecoﬂexltalc (70/30) blend ......................... 133
Table 4-1: Extrusion Processing conditions for the synthesis of Ecoﬂex-TPS graft

copolymers .............................................................................................................. 158
Table 4-2: Extrusion processing conditions for the reactive maleation of starch using

maleic anhydride/maleic acid modiﬁer in the presence of glycerol plasticizer ..... 162
Table 4-3: Extrusion processing conditions in the synthesis of Ecoflex-MTPS graft

copolymers .............................................................................................................. 164

Table 4-4: Blown Film Processing Conditions for Polyester-Starch graft copolymers.. 167

Table 4-5: Soxhlet Extraction results of Ecoflex-TPS graft copolymers ....................... 171

xxi

Table 4-6: Soxhlet Extraction results of Ecoflex-MTPS graft copolymers .................... 191

Table 4-7: Tear and Puncture properties of Ecoflex-MTPS (70/30) graft copolymer.... 204

xxii

TABLE OF ABBREVIATIONS

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Chapter 1
ATSB Aluminum tri-sec butoxide
CL Caprolactone
FT IR Fourier Transformed Infra Red Spectroscopy
LEcoﬂex Cross-linked Ecoﬂex
LPCL Cross-linked PCL
MTPS Maleated Thermoplastic Starch
PCL Poly (e-caprolactone)
PCL-co- Poly (e-caprolactone)-co— Poly (1,4-dioxan-2-one)
PPDX
PDX 1,4-dioxan-2-one
PFR Plug Flow Reactor
PI-IB Poly hydroxy butyrate
PHBV Poly hydroxy (butyrate-co-valerate)
PLA Poly (D,L, or DL-lactide)
PPDX Poly (1,4-dioxan-2-one)
ROP Ring Opening Polymerization
Chapter 2
CMR CL mass rate
DEG Diethylene glycol
DSC Differential Scanning Calorimetry

 

 

xxiii

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DP Degree of Polymerization
GMA Glycidyl methacrylate
ISR Initiator Solution mass rate
KOH Potassium hydroxide
MHS Mark-Houwink-Sakurada
MMR Monomer Mass Rate
PET Polyethylene terephthalate
PMR PDX mass rate
PVC Poly vinyl chloride
REX-PPDX Reactive Extruded Poly (1,4-dioxan-2-one)
RPM Revolutions per minute
RTD Residence Time Distribution
SAN Styrene Acrylonitrile copolymer
SEC Size Exclusion Chromatography
TGA Thermal Gravimetric Analysis
TSE Twin-screw extruder
X Monomer conversion
Chapter 3
ESEM Environmental Scanning Electron Microscopy
LLDPE Linear low density polyethylene
Luperox 2,5-Bis (tert-butylperoxy)-2,5-dimethylhexane
101
THF Tetrahydrofuran
\
xxiv

 

 

 

Chapter 4

 

 

 

 

 

 

 

 

 

 

DCM Dichloromethane
DMSO Dimethyl sulfoxide
DSR Destructurized starch
EAA Poly (ethylene-co-acrylic acid)
Ecoﬂex Poly butylene (adipate-co-terephthalate)
EVOH Polyethylene-vinyl alcohol
MA Maleic anhydride
SAXS Small angle X-Ray Scattering
TPS Thermoplastic Starch
WAXS Wide angle X-Ray Scattering

 

 

XXV

 

Chapter 1

 

INTRODUCTION

 

1.1 RATIONALE

During the last century, a tremendous advance in the ﬁeld of materials has been made
with the introduction of plastics. In today’s world, life without plastics is
incomprehensible. While plastics are strong, lightweight, inexpensive and easily
Processable, they are not readily broken down by the natural elements in the
environment. This is of particular concern when plastics are used in single-use disposable
packaging and consumer goods. Thus, new products have to be designed and engineered

from “conception to reincarnation” incorporating a holistic “life cycle thinking

 

approach”. This has opened up new market opportunities for developing biodegradable
and biobased products as the next generation of sustainable materials that meets
ecological and economic requirements (1-3).

The use of annually renewable biomass as opposed to petrochemicals as the
feedstocks for the production of polymers, chemicals and fuel needs to be understood
from a global carbon cycle basis. Clearly, petrochemical feedstocks are also natural.
Their fossil resources are formed from biomass over geological time frames. However,
the rate at which biomass is converted to fossil resources are in total imbalance with the
rate at which it is consumed. The use of annually renewable biomass to produce polymer
materials, chemicals, and fuel as an adjunct to fossil resources, would begin to move the
rate of carbon dioxide ﬁxation more in balance with the rate at which it is released.
Furthermore, if we manage our biomass resources effectively by making sure that we
plant more biomass than we utilize, we can begin to start reversing the carbon dioxide
rate equation and move towards a net balance between carbon dioxide
ﬁxation/sequestration and release due to consumption (4).

Environmental regulations, societal concerns, and a growing environmental
awareness throughout the world have triggered a paradigm shift in industry to develop
Products and processes compatible with the environment. In this regard, there is an
interest in developing and evaluating materials based on recyclable polymers, natural
POIYmers and degradable polymers. These include natural polymers like starch, cellulose,
PTOteiIIS, synthetic biodegradable polymers like poly (e-caprolactone) (PCL), poly (D, L,
or DL'laCIide) (PLA), poly hydroxy butyrate (PHB), poly hydroxy (butyrate-co-valerate)

(PHBV). P01)’ (1, 4.dioxan-2-one) (PPDX) and their modiﬁed versions. Along with the

 

above mentioned polymers, the entire portfolio of biodegradable plastics are used in a
variety of applications ranging from packaging ﬁlms, cutlery items such as spoons,
knives, forks etc., biodegradable packaging foams and many more.

Extruders are the most common machines in the Plastics Processing Industry.
Extruders have been used, in the traditional sense, to transform solid plastic into uniform
melt, for delivery to the next stage of processing. Various reactors (Batch, PFR, CSTR or
a combination of these) have been used to conduct chemical reactions that include
polymerizations, polymer functionalization, reactive grafting etc. The concept of using
extruders as reactors is novel and unique and is termed as “Reactive Extrusion”. More
Speciﬁcally, reactive extrusion refers to the process of conducting chemical reactions
during the melt extrusion process. There are reports (5-10) in the literature on reactive
extrusion processes. The reaction times needed to achieve near complete conversion need
to be well within the residence times available in extrusion operations (typically less than
5 rrrinutes). The advantage of using extruders to conduct reactions is as follows:

0 Fast and continuous process
0 Efﬁcient devolatilization capability through the vent port.

0 Modular in design and hence easy to scale up.

Based on the above rationale, the overall goal is to study and evaluate three diﬂferent
reaction chemistries such as Ring Opening Polymerization, Polymer cross-linking and
Reactive grafting of synthetic biodegradable and natural polymers in extruders. This
novel reactively extruded biodegradable biobased polymeric materials and blends ﬁnd

use in biodegradable plastics technology for ﬁlm applications.

 

1.2 PROPOSED GOALS

This study targeted the design and engineering of a new process for the extrusion
polymerization of 1, 4-dioxan-2-one (PDX) monomer, the co-polymerization of PDX
monomer with e-caprolactone (CL) monomer and the characterization of the resulting
polymers. It further deals with the reactive extrusion cross-linking and characterization of
biodegradable polyesters such as PCL (aliphatic polyester formed by the Ring Opening
Polymerization [ROP] of CL monomer) and Ecoﬂex (aliphatic-aromatic co-polyester
based on the monomers adipic acid, terephthalic acid and 1, 4-butane diol). It ﬁnally
targets the synthesis of Ecoﬂex-starch graft copolymers by reactive extrusion processing
and the subsequent characterization of the graft copolymers. Speciﬁcally, the study
focused on the following:

0 Reactive extrusion Ring Opening Polymerization (ROP) of PDX and co-

polymerization with CL monomer.
O Reactive extrusion cross-linking of PCL and Ecoﬂex.

0 Ecoﬂex-Starch graft copolymers by reactive extrusion processing.

 

1.3 SPECIFIC OBJECTIVES

o Reactive extrusion Polymerization of PDX and co-polymerization with
CL monomer.

. Bulk (solvent-free) homo-polymerization of PDX monomer
and co-polymerization with CL monomer using Aluminum
tri-sec butoxide (ATSB) as a catalyst in a batch reactor.

' Development of an extrusion polymerization process to
synthesize three—arm PPDX and Poly (8-caprolactone)-co-
Poly (1, 4-dioxan-2-one) (PCL-co-PPDX) using ATSB as a
catalyst.

' Characterization of PPDX and PCL-co-PPDX polymers
produced by the bulk extrusion polymerization process-
Molecular weight, Thermal Analysis.

' Modeling reactive extrusion polymerization of CL
monomer in a BAKER PERKIN extruder as an axially

dispersed Plug Flow model.

0 Reactive extrusion cross-linking of PCL and Ecoﬂex.
I Development of an extrusion process to cross-link PCL and

Ecoﬂex using free radical initiation chemistry.

 

Characterization of the cross-linked PCL (LPCL) and
cross-linked Ecoﬂex (LEcoﬂex) - Gel fraction and Thermal
Analysis.

Reactive extrusion blending of Ecoﬂex and their cross-
linked versions with talc.

Characterization of the blends- Thermal analysis and

mechanical properties of blown ﬁlms.

- Ecoﬂex-starch graft copolymers by reactive extrusion processing.

Ecoﬂex-starch graft copolymers by reactive extrusion
processing using maleic acid as a trans-esteriﬁcation
catalyst.

Characterization of the graft copolymers- Soxhlet
extraction, Fourier Transformed Infra Red Spectroscopy
(FTIR) analysis, Thermal Analysis and Mechanical
Properties of ﬁlms.

Development of an extrusion process to maleate starch in a
twin-screw extruder using maleic anhydride or maleic acid
to form maleated thermoplastic starch (MTPS).
Characterization of MTPS- Fourier Transformed Infra Red
Spectroscopy (FTIR), Thermal Analysis and Intrinsic

viscosity measurements.

 

' Reactive blends of PCL and Ecoflex with MTPS in
extruders.

' Characterization of the blends- Soxhlet extraction, Thermal
Analysis, FTIR Analysis and Mechanical Properties of

ﬁlms.

1.4 STRUCTURE OF THESIS

The thesis is composed of several chapters, each of which individually addresses the
work that has been done in relation to the speciﬁc objectives outlined above.

Chapter 2 deals with the bulk extrusion polymerization of PDX monomer and the co-
polymerization of PDX monomer with CL monomer to produce high molecular weight
PPDX and PCL-co-PPDX polymer. Chapter 2 also deals with particular reference to
screw conﬁguration, residence time requirements, molecular weight of the polymer ‘
produced, thermal analysis and modeling experiments. Chapter 3 explains the reactive
extrusion modiﬁcation of biodegradable polyesters by cross-linking. Characterization
methodologies with particular reference to gel fraction and Percent Crystallinity are
discussed. Chapter 3 also deals with blends of the biodegradable polyesters and their
cross-linked versions with tale for blown ﬁlm applications. Chapter 4 deals with
Synthesis and characterization of Ecoﬂex-starch graft copolymers. Chapter 4 also deals
With the synthesis of Maleated Thermoplastic Starch (MTPS) and Ecoflex/MTPS graft

COPOIymers by reactive extrusion, and its subsequent characterization using Thermal

 

analysis, FTIR, and mechanical property measurements. Summary and Conclusions are

outlined in Chapter 5 and Recommendations for future work are discussed in Chapter 6.

 

Chapter 2
\

BULK EXTRUSION POLYMERIZATION 0F 1, 4-DIOXAN-2-ONE
\

2-1 INTRODUCTION

1 ’ 4~dioxan-2-one (PDX) is a six-member lactone that is conveniently prepared by the
OXi ClEltive cyclization of diethylene glycol (DEG) (11). Figure 2-1 represents the oxidative
eye liZation of DEG to PDX with the competing dehydration reactions. PDX is a useful
In<)l"01ner and co-monomer for the synthesis of biodegradable aliphatic polyesters.
Pol ymerization occurs by a consecutive ROP to give high molecular weight polymer. The

h-
1 gh polarity and reactivity of PDX make this lactone a useful chemical intermediate and

s’O“’ent.

Diethylene Glycol 1,4-Dioxan-2-ol

1).—{Fri

OH HO
p-Dioxanone
H20 ‘H20
0 [0:0
O O
1 ,4-Dioxane 2,3-Dihydro—1,43Dioxin/Dioxene

Figure 2-1: Oxidative cyclization of DEG to PDX and competing dehydration

reactions

The properties of PDX are listed in Table 2-1.

Table 2-1: Properties of PDX monomer

 

 

 

 

 

Melting Point 27°C
Boiling Point 212-214°C @ 760mm Hg
Density @ RT 1.216 g/ml
Flash Point 255°F; 123°C
Molecular Weight C4H¢O3 102.09

 

 

 

 

10

The chemical reactivity, solvent properties, and high flash point make PDX useful in

several applications. The patent literature contains multiple uses for PDX; Table 2-2

summarizes patented uses for PDX.

The structure of PDX monomer is depicted in Figure 2-2.

O

O O

Figure 2-2: PDX monomer

ll

.l_ |,i

Table 2-2: Patented PDX end uses

 

 

 

 

 

 

 

 

 

 

US. Patent Assignee Application
Number
2,631,989 Am. Cyan. Solvent for Acrylonitrile Polymerization
(1953)
3,280,065 Dow Latex Stabilization
(1966)
3,631,189 Eastman PDX salts Useful Stabilizers For Halogen
(1971) Containing Polymers (i.e., PVC)
2,803,646 Jefferson PDX Ether-Esters Plasticizers and Lube Additives
(1957) Chem. Co.
3,268,310 Dow PPDX Controlled Dustiness and Improved Ignition
(1966) of Charcoal Briquettes.
3,294,743 GAF Epoxy Ether Resin PDX Compositions
(1966)
3,351,485 Dow Preserving Wood By PDX Irnpregnation
( 1967)
¥ 3 .264,182 Dow A Process for the Preservation of Specimens of the
( l 966) Animal Kingdom
\

 

 

 

 

12

 

The primary reactivity site on the PDX molecule is the ester carbonyl. The carbonyl is
reactive toward nucleophiles such as water, alcohols, and amines. These nucleophiles are

known to react with PDX to afford the corresponding ring opened product (Refer Figure

2-3).

0 o 0
+ H—X—R —> R\ JK/Ow
x OH
0
O JOK/
0 0 n o
R o \
:r “0”, \Ok/ \/\ RC”, 0 \/\on
0

OH +120

Figure 2-3: Reactivity of PDX monomer

Reaction of PDX with excess alcohol catalyzed by strong acid produces the
Corresponding ester ether. This ﬁnal reaction product proceeds through the intermediate
hydroxyl ester (12). Ring opening polymerization (ROP) of PDX monomer gives poly (p-
diox anone) (PPDX), a biodegradable thermoplastic poly (ester-alt-ether). Both in vivo
and in vitro, PPDX has good biodegradability. PPDX can degrade in vivo, and the
degrading products of each step are in accord with the normal metabolites of the body.
MOSt of them are excreted via the respiratory tract, and the reminder is extracted via the
alimevntary tract (13). The hydrolytic degradation of PPDX is a consequence of the

breaking of the ester bonds in every repeating unit of PPDX. This is generally studied in

13

a phosphate buffer solution of pH 7.4 and a temperature of 37°C. It is reported that the
increase in degree of crystallinity in the beginning of hydrolysis may be attributed to the
ester bond breaking ﬁrst in the amorphous region and then in the crystalline region (14-
16). Nishida et al. have also demonstrated that PPDX is decomposed to carbon dioxide

and water by microorganisms isolated from natural environments (17).

2.1.1 Mechanism of Polymerization of PDX monomer

The ROP of lactones and lactides has been thoroughly investigated during the last 40
years, due to its versatility in producing a variety of polymers in a controlled manner.
There are two main reasons for studying the polymerization of cyclic esters. The primary
reason is to exploit the potential of synthetic polymer chemistry to prepare a variety of
polymers with control of the major variables affecting polymer properties. A second
reason for studying ROP is to enable various advanced macromolecules, including
homopolymers with well-deﬁned structures or end groups, to be prepared, as well as co-
POl ymers with different architectures, e. g., block, graft, or star co-polymers (18).
ROP of lactones begins by addition of the lactone to the initiator. Chain
Propagation continues by consecutive addition of lactone to the end of the growing

P01 ymer chain. Chain propagation is described in Figure 2-4.

14

O
06: JOKMM—
R—O OH

R—OH ———->

O

 

O O
k0) \ etc
R—O OH

Figure 2-4: Chain Propagation in Lactone Polymerization

Bach macromolecule formed generally contains one chain end terminated with a
functional group originating from the termination reaction and one tenninus end capped
with a functional group originating from the initiator. Using different catalysts or
initiators, the nature of the functional groups can be varied to ﬁt the desired application.
Moreover, the ROP can be performed either as a bulk polymerization, or in solution,
emulsion, or dispersion (19-23). Depending on the initiator, the polymerization proceeds

according to three different major reaction mechanisms, viz., cationic, anionic, or

C‘
coordination-insertion” mechanisms (18)-

15

The cationic ROP involves the formation of a positively charged species, which is
subsequently attacked by a monomer. The attack results in ring opening of the positively

charged species through an SN2-type process. Figure 2-5 depicts the reaction mechanism.

O

P+ *ﬁ JCT—Uh _» P/OYRVO+\__'

0

Figure 2-5: Cationic ROP Reaction Mechanism

Anionic ROP of cyclic ester monomers takes place by the nucleophilic attack of a
negatively charged initiator on the carbonyl carbon or on the carbon atom adjacent to the
acyl oxygen, resulting in linear polyester. The propagating species is negatively charged
and is counter balanced with a positive ion. Depending on the nature of the ionic
prepagating chain end and the solvent, the reacting complex varies from completely ionic
to almost covalent. Figure 2-6 demonstrates the reaction mechanisms for the ROP of a

C yc lic ester by anionic initiation.

16

M +
0
ML". “i
2 R

Figure 2-6: Anionic ROP Reaction Mechanism

0
+ (I) «'1' R
\
2
2

In the case of reaction by coordination-insertion mechanism, the propagation is thought
to proceed by coordination of the monomer to the active species, followed by insertion of
the monomer into the metal-oxygen bond by rearrangement of the electrons. Figure 2-7

depicts a schematic for the coordination-insertion mechanism.

0

O\\ 0
\ . RO----C—Fl' ll
M—OR + R —-> I l |—>\M—o/\R'—c—0R
/ O—i /

_M.--- 0—

Figure 2-7: Coordination-Insertion Reaction Mechanism

The growing chain remains attached to the metal through an alkoxide bond during the
proPagation. The reaction is terminated by hydrolysis forming a hydroxyl end group. It

has been shown that the ROP of PDX to form PPDX polymer proceeds via the

17

coordination-insertion mechanism (24). Figure 2—8 represents the coordination-insertion

polymerization mechanism to form high molecular weight PPDX chains.

/OR
3
O O «it“‘OR O O—N\
s" on
a?
+ Flo—AI : :
0d \OR 0d O\/1

< ﬁ /OR
\J OR ——> RO—C- CHz—O—CHZ-CH2—0_N\

OR

0
("-1)

OJ
OR

RO C-CHz-O-CHZ- CHz—O Al<
OR

 

V
O

n
Figure 2-8: Coordination-Insertion Mechanism in the formation of PPDX polymer

Lactone polymerization can be accompanied by two common side reactions;
mtetmolecular and intramolecular trans-esteriﬁcation. These two reactions are illustrated

in Figure 2-9 for PPDX. Intermolecular trans-esteriﬁcation occurs when the hydroxyl end

18

group undergoes trans-esteriﬁcation with a different polymer chain. This process
broadens the molecular weight distribution of the polymer until the equilibrium value of
two is reached. Intramolecular trans-esteriﬁcation produces cyclics, by ‘back biting’ of
the hydroxyl end onto the same polymer chain. This reaction is of utmost importance, for
if the concentration of cyclics becomes too high, polymer properties and processing may
be adversely affected. The focus of this work was on the bulk extrusion ROP of PDX

monomer using aluminum alkoxides initiators, speciﬁcally aluminum tri-sec butoxide.

Intermolecular M
R- -—O LOH

 

 

 

 

 

 

 

VOW + R—OH

19

Intramolecular

R— ojK/Oviojk/OWOJOK/OA H

 

O

:OJK/Ov :O
Coat,

Figure 2-9: Intermolecular and Intramolecular trans-esterification of PPDX

+ R—OH

 

 

2.1.2 Lactone Polymerization Catalysts

There are numerous catalysts available in patents for the polymerization of lactones.
Union Carbide had several broad-based cyclic ester polymerization patents, which are
valid, but expired (25). GAF was ﬁrst to patent polymerization catalysts speciﬁc for PDX
pOlymerization. These patents fall into two classes: 1) Alkyl, alkoxy, and hydrido
aluminum catalysts (26); 2) Alkyl, alkoxy, and halide complexes of group IIB transition

20

metals (27). These catalysts showed good activity between 0 and 45°C at catalyst
concentrations of around 0.01 mole%. Dow patented the use of 1, 3-diketone complexes
of iron, titanium, and zinc catalysts speciﬁcally tailored for PDX polymerization (28).
Unlike the earlier case, the activities of these catalysts were low requiring about 0.2-0.5
wt-%. Polymerization temperatures also ranged from 60-90°C and gave low molecular
weight PPDX polymers(< 5000 g/mol). Eastman Kodak (29) also patented the use of tin
(IV) complexes to produce low MW PPDX polymers. Ethicon has the largest number of
PDX/PPDX patents on polymerization, processing and fabrication. Ethicon’s focus is
primarily on the medical device market. Ethicon’s earlier patent literature described PDX
polymerization in the bulk below the melting temperature of PPDX polymer (30).
Recently, Ethicon’s patents are focused on PDX melt polymerization, which allows the
preparation of a wide range of copolymers and gives uniform molecular weight
distribution (31-35). More recently, there has been considerable work in using zinc-based
catalysts, speciﬁcally Zn-L-lactate, in the bulk polymerization of PDX to form PPDX
polymer. While the highest molecular weights were obtained at 100°C, long reaction
times between 8 to 14 days were required to obtain optimum results (36). Nishida et a1.
(37) investigated the PDX polymerization in bulk with Sn (II) ethylhexanoate or
triethylaluminum over a range of 60-180°C. More recently, Raquez et a]. have studied
PDX melt polymerization, as initiated by aluminum triisopropoxide (Al (OH) 3) and
compared to the ROP promoted by tin (H) octoate (24). Nishida et al. have conducted the
emlb’matic polymerization of PDX monomer with 5-wt% immobilized lipase CA. The
POIYIDerization was carried out at 60°C for 15 hours and the highest molecular weight

obtained was 41,000 g/mol (38).

21

2.1.3 Kinetics of PDX Polymerization

PDX bulk polymerization kinetics has been investigated in a wide variety of conditions
using different catalysts as explained above. It has been shown that zinc lactate initiated
polymerizations are relatively slow when compared to Sn (ID—ethylhexanoate
(Sn (Oct) 2). However, it has the advantage that it does not catalyze backbiting
degradation even at reaction times of 14 days. Also, zinc chloride was shown to be less
effective for the polymerization as compared to zinc lactate and zinc bromide. Yet, for
any preparative application, zinc lactate has two advantages: Firstly, the lactate anion is
part of the human metabolism in contract to the bromide ion, and secondly, zinc lactate is
less hygroscopic, and thus more stable on storage than zinc bromide. Kricheldorf et al.
also found that bulk polymerizations of PDX at 100°C with zinc lactate, as a catalyst
never exceeded 80% yield (26). Although the polymerization rates using the three above-
mentioned initiators were distinctly different, the equilibrium monomer conversions were
almost the same at each temperature. The equilibrium conversions decreased in a similar
manner with increasing polymerization temperature. These results indicate that no matter
which initiator is used for the polymerization, the conversion of PDX monomer always
converges to a thermodynamic equilibrium. The kinetics of propagation for an

equilibrium polymerization system is generally expressed as follows in equation (2.1).

ll”[1441. —[M].)

__ (2.1)
[M]. - [M].

kw:

1“ the equation shown above, [M 10 is the initial monomer concentration, [M L is the

equilibrium monomer concentration, [M 1, is the actual monomer concentration at time t

22

and kw is the apparent rate constant. In all cases, a linear time dependence of

ln{{[M 10 — [M L} / {[M I, — [M L}} is observed after an induction period close to 6
minutes. This induction period could be attributed to the formation of active initiating
species.

When the polymerization was conducted in the presence of aluminum tri-
isopropoxide (Al (OiPr) 3), for similar initial monomer to initiator ratios, the
polymerization is faster. Also, PDX conversion, for example, increases from 60% to 80%
when Sn (Oct) 2 is replaced by Al (OiPr) 3 for an initial monomer to initiator ratio of 100.
Even here, it has been determined that the time dependence of polymer conversion passes
through a maximum before leveling off at the equilibrium monomer concentration. Table
2-3 compares the effect of using Sn (Oct) 2 and Al (O’Pr) 3 on the yield of PPDX obtained

in bulk at 100°C.

23

Table 2-3: Influence of the nature of the initiator on the yield of PPDX obtained in

 

 

 

 

 

 

bulk at 100°C.
Entry Initiator [DX] Polymerization time (min) Polymerization yield (%)
o/metal
1 Al (O‘Pr) 3 1000 270 80
2 Sn (Oct) 2 100 270 60
3 Al (O‘Pr) 3 300 270 84
4 Sn (Oct) 2 300 270 71
5 Al (O'Pr) 3 1000 375 70

 

 

 

 

 

 

 

Furthermore, kinetic experiments on PDX polymerization initiated by Al (O’Pr) 3 have
been conducted at 80°C and 100°C for various monomer to initiator ratios. It has been
detennined that for an initial [PDX L, / [Al] ratio of 600 at 100°C, the time dependence
of monomer conversion passes through a maximum after about 7 min, and then the
monomer conversion decreases and levels of at a value close to the equilibrium monomer
conversion at about 12 min. Moreover, it has also been shown that at 80°C, as the initial
monomer to initiator ratio increases, the time taken for the conversion to reach the
equilibrium value also increases. It has also been noted that a linear time dependence of

lniiIML, - [M L }{IM 1, - [M L }} is observed after an induction period close to 6 min. This

linearity attests that the bulk polymerization kinetics of PDX monomer initiated by Al
(Oin) 3 is ﬁrst order in monomer in the overall range of monomer conversions
inVEBStigated. Such a linear dependence also indicates that the number of growing chains

does not appreciably change during polymerization. From the slope of these linear

24

relationships, the values have been determined to be 0.293, 0.144 and 0.104 min'1 for
initial monomer to initiator ratios of 200, 400 and 600 respectively at 80°C (39). Figure
2-10 shows the effect of polymerization temperature on the equilibrium monomer
conversion for a [PDX lo / [Al] ratio of 600 and a polymerization time of 14 hours. It is
evident from the ﬁgure that equilibrium monomer conversions decrease with increasing

temperature.

1001
901
80d

701

 

501

401

304

Equlllbrlum monomer converclon (wt%)

10-

 

 

o i V V
0 50 100 150 200

Polymerization temperature (°C)

Figure 2-10: Effect of Polymerization Temperature on the monomer conversion
l’eaChed at equilibrium for a monomer to initiator ratio of 600 and a polymerization

time of 14 h.

25

Thus, PDX polymerization leaves high amounts of unreacted monomer due to a low
ceiling temperature. The micro reversibility model can adequately describe such

monomer-polymer equilibrium. According to the model, all growing chains
(M;andM;+,) are capable of depolymerizing until they reach a constant equilibrium

monomer concentration as explained by equation (2.2) and (2.3).

M i +M —"‘->M 1.. (2.2)
M;+l—-ﬂ—)M;+M (2.3)

Where k P and k dp correspond to the propagation rate constant and the depolymerization

rate constant respectively. US 5,717,059 describe a method for removing the PDX
monomer from previously solidiﬁed polymerizing mixture without any adverse
degradation reactions to form high molecular weight polymers. However, the time for
removing the unreacted monomer from the polymerizing mixture is in the order of hours.
Thermally stabilized PPDX polymers have also been prepared by end-capping the
extremity of PPDX in the melt. However, after end-capping of the PPDX extremity by a
chemical agent such as pyromellitic anhydride, the reacting mixture still contains a large
amount of unreacted monomer, which has to be removed from PPDX and recycled,

making the process economically unattractive (US No 5,652,331).

26

2.1.4 Thermodynamics of Polymerization

In order for a polymerization reaction to yield high molecular weight polymer at a given
temperature and pressure, the monomer-polymer equilibrium must be favorable to drive
the reaction forward and the reaction must be mechanistically feasible (40). All reactions
are generally characterized by a thermodynamic equilibrium as explained above in
equations (2.2) and (2.3). This equilibrium is characterized by equilibrium constant,
which to a ﬁrst approximation is equal to the inverse equilibrium monomer concentration

as explained in equation (2.4) and equation (2.5).

 

 

K: [Min] z 1 (2.4)
(1M3 [Mr] M

ln[M]_ = AG = AHO — A50 (25)
RT RT R

Equation (2.5) discloses the connection between the equilibrium monomer concentration
and the free enthalpy of polymerization as the driving force of the polymerization. For
standard conditions (initial monomer concentration equal to 1M) and for a certain critical
temperature, the equilibrium constant may be 1 and hence the equilibrium monomer
concentration is 1M. Under these conditions, no polymer is obtained. From a
thermodynamic point of view, the critical temperature is deﬁned as the ceiling
temperature (T c) with both AH and AS<0 (and as the ﬂoor temperature, Tf for both AH
and AS>0). For every polymer system, above Tc polymerization is impossible and
propagation occurs only if concentration of monomer is greater than the equilibrium

monomer concentration at a given temperature (41).

27

Monomer-polymer equilibrium is not the only equilibrium observed in ROP.

There is a second equilibrium to be considered, ring chain equilibrium.

M ; <——) M ;_x + R, (2.6)

As indicated by equation (2.6) the equilibrium constant of a certain cyclic Rx with degree
of polymerization x is to a ﬁrst approximation equal to the equilibrium concentration of
the respective cycle. According to Jacobson and Stockmayer (42), the equilibrium
constant is determined by and is proportional to the degree of polymerization to the —2.5

power and indicated by equations (2.7) and (2.8).

12.4114.-.) [4:1/ [M;l=[R.l=[-(331’V5]% I r 1 f2 (2.7)

 

.(ZNX) < r: >

K (2.8)

.. _1__.
x x%
with [R,] being the equilibrium concentration of the cyclic oligomers with degree of

polymerization x, N the number of bonds of the oligomers, and < r],2 > the mean square

radius of gyration of a polymer segment with degree of polymerization x.
Thermodynamic data for small and medium sized lactones show that the entrOpy
change during polymerization is negative. Thus, the driving force for polymerization is
the negative change of entropy. The ROP of highly strained three-and four-membered
rings is a favorable reaction- the driving force being the release of angular strain (Bayer’s
strain). The presence of substituents at the ring carbons further increases the strain and
thus increases the exotherrrricity of the reaction. In medium sized rings, such as a 7-

membered ring, the relief of intramolecular crowding (transannular strain) is the driving
force. The A03 values for B-propiolactone, y—butyrolactone, 8-valerolactone,

28

E-i

PI)

‘10

e-caprolactone, and (1,0t-dimethyl—B-propiolactone have been reported as -60.2, +12.6, -

8, -12.8, and -84.02 kJ/mol, respectively. The AG: is positive for y—butyrolactone,

indicating that the ROP of 'y—lactone is thermodynamically unfavorable because of the
stability of the 5-membered ring (43,44). However, by use of drastic reaction conditions,
polymerization may be carried out. For example, poly (y-butyrolactone) of Mn 35,000
g/mol could be prepared under a pressure of 20,000 atrn at 160°C (45,46).

ROP of lactones begins by addition of the lactone to the initiator. Chain
propagation continues by consecutive addition of lactone to the end of the growing
polymer chain. The conformational strain of six member ring cyclic esters, such as PDX,
differs from other six member heterocyclic rings not containing a carbonyl group such as
tetrahydropyran or 1,4-dioxane etc. These cyclic ethers do not undergo ROP due to the
stability of the ring conformation adopted as explained above. With six member ring
lactones, ring strain is induced from the planar conﬁguration of the ester group. In
addition, this ring strain twists the planar ester group, causing resonance destabilization
(47). Both these factors result in the polymerizability of six member ring lactones such as
PDX.

There is limited work done on the thermodynamics of PDX polymerization. The
equilibrium monomer concentration is related to temperature using Dainton’s equation as

shown below.

 

k AH
RlnlM L = Rln[kl] = T P _ AS: (2.9)

P

In equation (2.9), AH p is the enthalpy of polymerization under the prevailing

experimental conditions, AS: is the entropy change at the standard state

29

([ML =1mol/ L) and R is the perfect gas constant. By doing a linear regression analysis
and plottingRlnlM L vs; , the enthalpy of polymerization is determined to be -14.1

KJ/ mole and the entropy of polymerization is detennined to be -45.3 J / mole.K . The

enthalpy of polymerization reported is less negative than that reported for L—lactide. This
may be attributed to less ring strain for the non-substituted ether-ester PDX ring than for
the substituted diester L-lactide ring (39). The thermodynamic parameters mean that PDX
polymerization has a less negative free energy than that for L-lactide. These values are in
quite good agreement, with the values reported by Nishida et al. (38). The precipitation of
the polymer in the low temperature bulk polymerization of PDX profoundly inﬂuences
the monomer polymer equilibrium. For example, polymerization of trioxane to polyacetal
at or above the T m of polyacetal is equilibrium limited to less than 50%. To drive the
reaction, polyacetal is prepared in molten trioxane. Under the polymerization conditions,
polyacetal precipitates from the molten trioxane. Chain growth proceeds both in the
liquid phase and in the crystalline phase. The precipitation of polyacetal from solution
shifts the trioxane-polyacetal equilibrium, allowing complete conversion of monomer to
polymer. Similarly, the PDX to PPDX equilibrium is shifted by the precipitation of

PPDX from solution.

2.1.5 Molecular Weight and Molecular Weight Distribution

The two main polymerization parameters that determine the molecular weight of the

PPDX polymer are initiator concentration and monomer to polymer conversion. The total

30

initiator concentration determines the number of polymer chains for a given
polymerization reaction. The number of monomer units in each polymer chain is then
determined by the conversion of monomer to polymer. Therefore, the two-polymerization
parameters that determine the molecular weight of the PDX polymer is; initiator
concentration and the monomer to polymer conversion. Equation (2.10) shown below
provides the number average molecular weight as a function of the initiator concentration
and conversion of monomer to polymer.

Mn = [%Conv. * PDX (moles)/ Initiator(moles)]*102.09 (2.10)

While equation (2.10) predicts the number average molecular weight, it does not predict
molecular weight distribution, which is determined by speciﬁc polymerization conditions
(48).

Kricheldorf et al. polymerized PDX monomer using zinc lactate catalyst and
determined that 100°C was the optimum temperature for polymerization to obtain a high
molecular weight. Moreover, up to a monomer to initiator ratio of 4000, highest viscosity
values were obtained at the longest reaction times. These results again demonstrate that
zinc lactate is a relatively inefﬁcient catalyst, but it has the advantage that it does not
catalyze backbiting degradation. When zinc chloride was used as the catalyst, the
molecular weight obtained at a monomer to initiator ratio of 2000 was signiﬁcantly lower
than the value obtained with zinc lactate (36).

Nishida et a1. polymerized PDX at 80°C and determined that the number and
weight average molecular weights increased linearly with conversion. These results
suggest that this process is a living mechanism (37,49). However, since the

polymerization is an equilibrium polymerization, the polydispersity of the polymers

31

broadened to about 2; the most probable distribution of molecular weights, with increase
in conversion. Equilibration and depropagation probably occur by bimolecular
transesteriﬁcation as in the case of e—caprolactone reported by Kowalski et a1. (50) and
Schindler et a1. (51).

Raquez et a1. polymerized PDX monomer in bulk using aluminum triisopropoxide
as initiator. It was determined that with an increase in the [PDX lo / [Al] ratio, the PPDX
molecular weight (with reference to polystyrene standards) increases gradually with time.
Furthermore, there is a relative increase of Mn with an increase in the [PDXlo/ [Al]

ratio at constant monomer conversion. As an example, M,I passes from 6000 g/mol to
11,200 g/mol as the [PDX L, / [Al] ratio increases from 400 to 600, respectively. It has

also been observed that by using chloroform added with 0.3-wt% of tri-n-
octylmethylammonium chloride (T OAC) as an eluent at 40°C; low values of apparent
molecular weight are obtained as expressed with reference to polystyrene standards and
low polydispersities (MW/Mn < 1.3). This is due to the poor solubility of PPDX in
conventional organic solvents, which leads to the formation of very compact coils,
characterized by a low hydrodynamic volume compare to highly soluble polystyrene
(39).

Nishida et al. also polymerized PDX at 60°C for 15 hours with 5 wt%
immobilized Lipase CA as an enzymatic catalyst. The product without puriﬁcation
showed multimodal distribution and wide polydispersity. This wide polydispersity is
attributable not only to the heterogeneous polymerization but also to the simultaneous

degradation process (52,53).

32

2.1.6 Extrusion Studies

The traditional use of an extruder is to use heat, pressure, and shear to transform solid
plastic into a uniform melt, for delivery to the next stage of processing. This frequently
involves mixing in additives such as color concentrates, blending resins together, and
incorporating regrind. The extruder accomplishes all this by using a barrel, a hollow tube,
containing a screw with helical channels. The screw is generally divided into three
sections: (1) the solids conveying section, (2) compression section and (3) the metering or
pumping section. The major purpose of the feed section is to convey the plastic pellets
forward through the extruder, once they have been fed into the barrel at one end through
the hopper and feed port. In the compression section of the screw, the plastic is melted.
The depth of the screw channels decreases, building up more pressure on the plastic and
forcingthe solid against the barrel wall. External heating, as well as internal heating from
friction between the plastic and the barrel and especially from the shearing of the plastic
itself, causes the plastic to soften and melt. In the metering section of the screw, the
channel is again a constant depth. This part of the screw is responsible for delivering
molten polymer at a desired and uniform rate to the shaping device (54).

Reactive extrusion refers to the process of conducting chemical reactions during the
melt extrusion process. These reactions include polymerization, cross-linking, grafting,
chain extension etc. The advantage of using extruders to conduct reactions is as follows:

0 Fast and continuous process.

0 Reactions can be performed in bulk in the absence of solvent.

0 Efﬁcient devolatilization capability through the vent port.

33

 

0 Modular in design and hence easy to scale up.
However, in any reactive extrusion process, the following parameters are of key
importance

0 Screw Conﬁguration

0 Metering of reactants

0 Temperature proﬁle

0 Heat transfer requirements.
There are numerous references on grafting reactions conducted in extruders. Speciﬁcally,
grafting of functional groups like acids, esters and anhydrides have been studied by many
researchers (55-57). Cross-linking of polypropylene has been conducted in extruders
(58). Cross—linking of biodegradable polyesters will be disCussed in detail in the
forthcoming chapter. Chain extension of PET and that of polyarrrides have also been
investigated (59,60). Blends of synthetic polymers with natural polymers such as starch
and cellulose have been synthesized in twin-screw extruders (61,62). Table 2-4 represents

the various reactions performed in co-rotating twin screw extruders (63).

34

Table 2-4: Reactions performed in co-rotating twin screw extruders

 

 

 

Final Product Feed Product Type of reaction

Polyurethane Polyol + Diisocyanate + Polyaddition
Aromatic Diamine

Polyethylene Bis (Hydroxyethyl) Polycondensation

Terephthalate Terephthal ate

 

Potassium Aluminate

Clay + alkali solution

Salt formation

 

 

 

 

 

 

 

 

 

 

 

 

Polyarylate Bisphenol A + Phthalic acids Polycondensation
Polyoxymethylene Trioxane + comonomer Ionic Polymerization
IS Block Copolymer Isoprene, 1, 3-butadiene + Ionic block
styrene copolymerization
Polyarnide (Nylon 6) Caprolactam Ionic polymerization
SAN Copolymer Styrene + Acrylonitrile Ionic polymerization, free
prepolymer radical
Ethylene vinylacetate Polyethylene + vinylacetate Radical grafting
Terephthalic acid Isophthalic acid Isomerization
Polyalkylmethacrylate Methacrylate ester Radical polymerization
Polyols, Amines Polyurethane scrap Hydrolysis
Styrene monomer Polystyrene Depolymerization
Indoxyl Phenyl glycine Cyclization

 

 

 

35

 

2.1.6.1 Reactive Extrusion Polymerization

Reactive extrusion polymerization involves polymerizing a liquid/solid monomer or pre-
polymer within the residence time available in the extruder to form a high molecular
weight melt. The economics of using the extruder as a bulk polymerization reactor are
favorable when high throughputs and control of molecular weight are realized. This
places a limitation on the residence time required to complete the polymerization, which
ideally should be less than 5 minutes (64). Researchers have shown that extruders can be
used for bulk polymerizations of monomers like methylmethacrylate, styrene, lactam and
lactide (65-67). Further, there are various patents in literature describing the
polymerization of polyurethanes, polyacetals and polyetherimide in modular twin- screw
extruders (68-74). Since different polymerization mechanisms are involved in the
extrusion of these polymers, the process schematics and screw conﬁgurations vary
considerably. The extruder screw conﬁguration can have an effect on yield, molecular
weight, molecular weight distribution and product throughput.

Bulk polymerization of CL monomer in a co-rotating twin-screw extruder has
been investigated (75-78). Narayan et al. have used a reactive extrusion process for the
bulk polymerization of CL monomer using coordination insertion catalysts to form a high
molecular weight polymer at high throughputs (79-81). Bulk polymerization of CL
monomer using coordination insertion catalysts have been discussed by Young et a1 (82).
Narayan et al. (83) have also discussed bulk polymerization of CL monomer in an
extruder in the presence of starch to give a compatibilized blend of PCL, starch and

starch-g-PCL polymers. EP 626405 describes the preparation of low molecular weight

36

PCL polymer with Mll ranging from 35,000 g/mol to 66,000 g/mol at residence times
ranging from 6 to 15 minutes respectively. Low cost production and processing methods
for biodegradable plastics are of great importance since they enhance the commercial
viability and cost—competitiveness of these materials. Reactive extrusion is an attractive
route for the polymerization of a cyclic ester monomer, without solvents, to produce high
molecular weight biodegradable plastics. There is thus a deﬁnite need for an improved
bulk polymerization process for economically producing aliphatic ester polymers.
Compositions of such polymers that possess improved properties and processability are

useful as biodegradable articles, especially ﬁlms.

2.1.6.2 Screw Conﬁguration

Twin-screw extruders have been used for processing of viscous materials for over a
century. They are used for a wide variety of applications and are made in a number of
different geometries. The screw conﬁguration used in an extruder depends on the
application. There are three fundamental types of screw elements: forwarding, mixing
and zoning. The screw conﬁguration is generally limited to the torque handling capability
of the screw shafts.

Forwarding elements are used to drive and center mixers and zoning elements.
They also forward material past feed, vent, drain and output barrel holes. Forwarding
screw elements can be single flighted (monolobal), double ﬂighted (bilobal), or triple
ﬂighted (trilobal). Single ﬂighted screws have the largest channel cross-section and are

useful when large conveying capacity is required, such as in the feed section when

37

feeding low bulk density materials. Double ﬂighted screws have the second largest
channel cross-section. They are used when both good mixing and output is required.
Triple ﬂighted screws have the smallest channel cross-section and, consequently, the
lowest output capacity. They are useful when high shearing forces are required; triple
ﬂighted screws also yield high heat transfer capability (63). Left-handed screw elements
are primarily used for sealing purposes. A left handed screw element requires the
upstream screw section to be completely ﬁlled over a certain distance in order to generate
the pressure necessary to override the left handed section. Thus, one can intentionally
create a fully ﬁlled screw section to obtain complete sealing. This is frequently used just
before a vent port where vacuum is being applied.

Mixers will be dispersive, distributive or a combination thereof. They can also be
described additionally as having a pressure drop over them (restrictive), no pressure drop
(neutral) or a pressure rise over them (forwarding). Common distributive mixers include
narrow kneaders, slotted screw ﬂights, vane mixers, turbine mixers, slotted axial flow
elements, pin mixers, and the like. Common dispersive mixers include wide kneaders,
shaped lobals, radial barrier elements, axial barrier disc elements, large advance ﬂighted
solid elements, axial slotted restrictors with OD gaps to the barrels, etc. The strongest of
these dispersive mixers capture lobal and other regions from which material may be
elongationally acclerated. The most distributive mixers simply divide and recombine
materials with material domains remaining intact. Absence of lobal capture is often an
advantage for distributive mixing. Figure 2-11 represents the concepts of distributive and

dispersive mixing schematically.

38

Dispersive mixing

 

O

Distributive mixing

06)

V

   

Figure 2-11: Dispersive and Distributive Mixing schematic

A sequence could be described for a strong dispersive mixer of where material leaving
gentle screw channels to be captured in a lobal pool from which it experiences
elongational acceleration/extensional ﬂow/stretching and mixing, entering an overﬂight
region. Usually, when this condition is met, a productive elongational mixing condition
will also exist around the intermesh and apex regions (84).

Zoning elements are used to separate two sub-processes, such as a vacuum vent
from upstream screws, an injection/mixing sub-process from upstream and downstream
screws, ensuring the maximum size of a particle that can pass a given barrel point, and

the like. Most zoning elements also serve as. forwarding or mixing elements.

39

A typical screw conﬁguration in an extruder consists of conveying and kneading
elements. Conveying or forwarding elements are represented by a set of two numbers as
X/Y where X is the pitch of the element and Y, the axial length, both in millimeters
(mm). Reverse ﬂighted elements are represented as X/Y LH. Kneading or Mixing blocks
are represented as KB X/Y/Z where X is an angle that represents the degree of stagger, Y
is the number of kneading block elements and Z is the pitch in mm. Additionally, they
may be reverse staggered and represented as KB X/Y/Z LH (79). A typical screw

conﬁguration consisting of conveying and kneading elements is shown in Figure 2-12.

l

 

tannin

Anugggamngggnem
12°39 EQQ QEQ°\°\
auto brow HJir—NN
mmogggmmogggonomo
AAA\\\AAA\\\AAA
gwwmumhnmmumeEv—w
VVV\\\VVV\\\VVV
”RN HAr—e
{knee puss/,2:
A vv
p—e
EEV EEE
AA AAA
mil-D {ﬁr—s)—
vv vvv

Figure 2-12: Conventional screw configuration

2.1.6.3 Residence Time Distribution

The idea of using the distribution of residence times in the analysis of chemical reactor
performance was apparently ﬁrst proposed in a pioneering paper by MacMullin and
Weber. However, the concept did not appear to be used extensively until the early

1950’s, when Danckwerts gave organizational structure to the subject by deﬁning most of

40

the distributions of interest. In an ideal plug ﬂow reactor, all the atoms of material
leaving the reactor have been inside it for exactly the same amount of time. This time that
the atoms spend in the reactor is called the residence time of the atoms in the reactor. The
residence time distribution (RTD) of a reactor is characteristic of the mixing that occurs
in the chemical reactor. There is no axial mixing in a plug-ﬂow reactor, and this omission
is reﬂected in the RTD, which is exhibited, by this class of reactors. The CSTR is
thoroughly mixed and possesses a far different kind of RTD than the PFR. Thus, the RTD
exhibited by a given reactor yields distinctive clues to the type of mixing occurring
within it and is one of the most informative characterizations of the reactor.

The RTD is determined experimentally by injecting an inert chemical, molecule,
or atom, called a tracer, into the reactor at some time t = 0 and then measuring the tracer
concentration, C, in the efﬂuent stream as a function of time. In addition to being a non-
reactive species that is easily detectable, the tracer should have physical properties
similar to those of the reacting mixture and be completely soluble in the mixture. It
should also not adsorb on the walls or other surfaces in the reactor.

In a pulse input experiment, an amount of tracer Nois suddenly injected in one

shot into the feed stream entering the reactor in as short a time as possible. The outlet
concentration is then measured as a function of time. The efﬂuent concentration-time
curve is referred to as the C curve in RTD analysis. The quantity E (t) is called the
residence time distribution function. It is the function that describes in a quantitive
manner how much time different ﬂuid elements have spent in the reactor. It is deﬁned as

shown in equation (2.11).

41

5(2) = _°:£‘)_ (2.11)

IC(t)dr

0
The integral in the denominator is the area under the C curve. The principal difﬁculties
with the pulse technique lie in the problems connected with obtaining a reasonable pulse
at a reactor’s entrance. The injection must take place over a period, which is very short
compared with residence times in various segments of the reactor or reactor system, and
there must be a negligible amount of dispersion between the point of injection and the
entrance to the reactor system. E (t) is also called the exit-age distribution function. If we
regard the “age” of an atom as the time it has resided in the reaction environment, then E
(t) concerns the age distribution of the efﬂuent stream. The fraction of the exit stream that
has resided in the reactor for a period of time shorter than a given value t is equal to the

sum over all times less than t of E (t), or expressed as follows:
I E(t)dt = F(t) (2°12)
0

F (t) is deﬁned as a cumulative distribution function. Sometimes, the F curve is used in
the same manner as the RTD in the modeling of chemical reactors.

No matter what RTD exists for a particular reactor, ideal, or non—ideal, the
nominal holding time t, is equal‘to the mean residence time, tm. The mean value of the
variable is equal to the ﬁrst moment of the RTD function, E (t). Thus, the mean residence

time is as explained in equation (2.13)

42

th(t)dt .. (2°13)
rm = 3;———— = j tE(t)dt
I E(r)dt °

The second moment commonly used is taken about the mean and is called the variance,

or square of the standard deviation. It is deﬁned as follows in equation (2.14)

2

a = (t — tm )2 E(t)dr (2-14)

O‘--:8

The magnitude of this moment is an indication of the “spread” of the distribution; the
greater the value of this moment, the greater a distribution’s spread. The third moment is
also taken about the mean and is related to the skewness. The skewness is deﬁned as

follows in equation (2.15).

3 l
0.3/ 2

(herald: (245)

 

°‘——-o8

The magnitude of this moment measures the extent that a distribution is skewed in one
direction or another with reference to the mean. More recently, on-line measurements of
RTD in a twin-screw extruder (TSE) using simple Optical techniques have been
developed (85). Use of the RTD in screw design can minimize thermal degradation,
improve product uniforrrrity and control the residence time of the melt to allocate
sufﬁcient time to complete a reaction. The RTD, in combination with the reaction
kinetics, can be used with the assumption of segregated ﬂow to provide models for ﬁtting
reactive extrusion data without requiring complete understanding of the physical ﬂow

conditions (86).

43

2.1.6.3.1 Extrusion Modeling

Extrusion is one of the most versatile and energy-efﬁcient processes for the manufacture
of polymer products, including food, pharmaceuticals and plastics (87). Flow patterns,
mixing conditions and Residence Time Distribution (RTD) in any plasticizing and
cooking extruder have signiﬁcant effects on the extent of the thermal and shear treatment
provided to the extradites, and the extent of conversion and chemical reactions of the
biopolymers (88). Flow in the extruder has been modeled at various levels of
sophistication. A review of the basic features of extrusion models is available (89) and it
would appear that one could predict the gross features of the relationships among design
and performance variables with reasonable success. The major failures of simple extruder
models lie in two areas. Most models introduce simplifying assumptions with respect to
heat transfer between the extruder surfaces and the melt, and these assumptions are
probably rarely consistent with typical commercial operation. Furthermore, detailed
instrumentation of an extruder, from which heat transfer coefﬁcients might be obtained,
is rarely carried out and/or reported. The other major factor in the failure of simple
extruder models lies in the variety of screw designs. Simple extruder models include
simpliﬁcation with respect to screw geometry, and the nearly inﬁnite variability of
detailed geometrical parameters makes it impossible to consider more than a small
fraction of possible designs. Nevertheless, such models are useful for indicating how the
major variables are related, and so provide a rational basis upon which one may estimate

performance limits of a speciﬁc piece of equipment. In addition, such models should

provide some rational guidelines for scale-up from a small laboratory extruder-reactor to
a larger commercial system (90).

The ﬂow behavior of the twin-screw extruder (T SE) is also not well understood,
but is generally believed to fall between a CSTR (complete mixing) and a PFR (no axial
mixing). As explained in the earlier paragraph, model conﬁgurations such as PFR, CSTR,
PFR+CSTR, CSTR in series, and PFR+CSTR in series have been proposed for the ﬂow
proﬁle in a single-screw extruder. Few models have been proposed for the TSE because
its ﬂow proﬁles are much more complex. Bounie (91) used Fourier functions and
multivariable optimization techniques to predict a pulse response from several ﬂow
models, and reported that the model of PFR + nCSTR gave the best ﬁt to the
experimental data. Giudici et al. modeled industrial nylon-6, 6-polycondensation
processes in a twin-screw extruder reactor (92). Michaeli et al. (93) have developed an
extruder design based on a model that describes the mechanism of conveying, heat
transfer, and reaction in a co-rotating twin-screw extruder. The developed model has been
tested in practice on the production of homopolymers (nylon 6 and polystyrene) and a
copolymer (nylon 6 block copolymer). Zhang et al. (94) investigated the ‘living’ free
radical polymerization of butyl acrylate and styrene using a tubular reactor model.
Residence time distribution effects on polymer properties are studied by varying the

Peclet no (Pe,) to show the degree of back mixing that adversely affects polymer

properties. Polymerization of CL monomer has been studied in a co-rotating extruder and
modeled using kinetic and rheological data (78).
S. Hojabr et al. (95) modeled the melt grafting of glycidyl methacrylate (GMA)

onto polyethylene in a co-rotating twin-screw extruder as a modular reactor consisting of

45

plug ﬂow and axial dispersion reactor cells. There are also sources available in literature
where researchers have modeled the conveying elements of a screw as PFR and the
kneading blocks as a CSTR (96). Thus, it was suggested to model the bulk extrusion
polymerization process of CL monomer by an axial dispersed PFR model using RTD
analysis. Moreover, as a regular PFR model represented a good ﬁt of the batch process, it
was assumed that an axial dispersed PFR would represent a good ﬁt of the extrusion
polymerization process.

The axial dispersed PFR model can be thought off as a PFR superimposed with
some degree of back rrrixing or intermixing. This dispersion coefﬁcient, inversely

pr0portiona1 to Pe, can be determined by pulse tracer experiments as described in

equation (2.16).

0 _ 2 2 (l-e'Pe') (2.16)

Here 02 is the variance and tn, the mean residence time, evaluated from the experimental

RTD data. The cumulative F—curve F (2') can be obtained from equations (2.17) & (2.18).

r- 1’

 

 

 

(2.17)
C 1 l l- 2'
F(T)--C:—-2‘-Eelf ——'T—
u L
(2.18)

46

The predicted F- curve can be compared with the experimental F—curve to ﬁnd the
goodness of ﬁt (97). Knowing the kinetics of the reaction [equations (2.2) and (2.3)],

monomer concentration could be calculated by the axial dispersion model using equation
(2.19) and compared with the experimental values. In equation (2.19), C is the

normallzed concentratlon equal to E—, where C0 rs the 1n1t1al concentration of the
0

monomer; z is the normalized length of the extruder equal to —E-, where L is the total

extruder length; k pis the forward propagation reaction rate constant in lit/mol.min;

[11013 the inrtlal concentration of the initiator 1n mol Hit; 7 rs the mean res1dence time
u

in min obtained from the experimental RTD data and [C], is the equilibrium monomer

concentration at a given temperature.

1 d2 C _ dC + k,[1],LC = kp [1],,1,[c]¢ (2.19)

Fe, dz 2 dz

 

 

u uC0

Equation (2.19) can be solved using the boundary conditions represented in equations

(2.20) and (2.21).
z = 0; C = 1 (2.20)
‘ (2.21)
z = 0; LC = 0
dz

47

Equation (2.22) shows the solution to the above equation (2.19).

dz) = ('3) + exp(m, Z)[Ci 9030712 z) + c2 sin(m2z)] (2°22)

Expressions for the constants a, b, m1, m2, c1 and C2 are shown in equations (2.23)

through (2.28) respectively.

 

k, [1],,LPe, (2.23)
a = '—.—
u
mist m4)
Co
Per (2.25)
m, =
2
W (2.26)
m2 = ——
2
c. = 1 .. [2] (2.27)
a
,2 =_,l[!_"_] (2.28)
m2

48

2.2 EXPERIMENTAL DETAILS

The process requirements needed to develop an extrusion process to continuously

polymerize PDX monomer to form PPDX polymer are as follows:

Accurate metering of PDX monomer and initiator under an inert atmosphere of
nitrogen.

Adequate length of the extrusion zone to meet residence time requirements
needed to convert PDX monomer to PPDX polymer. However, in the case of
homo-polymerization of PDX, due to the dynamic chemical equilibrium between
the monomer and the polymer above the melting temperature of the polymer, it is
not possible to obtain complete conversion. When PDX monomer is co-
polymerized with a small amount of CL monomer (about 5-10 mol %), the length
of the extrusion zone becomes very critical.

Suitable screw conﬁguration in order to meet residence time requirements, melt
pumping needs and mixing requirements.

Good melt temperature control.

Reasonable product throughputs of the polymer.

Based on these requirements, a novel extrusion process was developed to continuously

polymerize PDX monomer. The details of the polymerization process developed to

homo-polymerize PDX monomer and co-polymerize PDX monomer with a few mole%

CL monomer in a twin-screw extruder will be outlined in the pages that follow. The

characterization of PPDX and PCL-co-PPDX polymers synthesized by the extrusion

process is also discussed.

49

2.2.1 Materials and Equipment

PDX monomer, provided by Shell Chemicals, was puriﬁed by vacuum distillation and
then dried over molecular sieves before the polymerization. The details of the distillation

and the drying step are provided in section 2.2.2. The initiator for polymerization was

 

aluminum tri-sec butoxide (ATSB), obtained from Aldrich Chemicals and used as
obtained. A 10 percent (w/w) solution of ATSB in anhydrous toluene (Aldrich
Chemicals) was used for initiation of the polymerization. The initiator solution was
prepared in a glove bag under an inert atmosphere of nitrogen. Extreme care was taken to
prevent the entrainment of ambient moisture in any of the process steps. CL monomer
was purchased from Dow Chemicals and was used as obtained. RTD studies were
conducted using a ﬂuorescent die, UV-XPBB obtained from LDP Net.

The polymerization of PDX monomer and its co-polymerization with CL
monomer was conducted in a Baker Perkins co-rotating Twin Screw (MPCN-30); 30
mm co-rotating, fully interrneshing screws were mounted in a barrel about 450 mm long.
The barrel sections were electrically heated and cooled by circulating water. The two-
strand hole die was heated using an external power supply. The screw elements needed

were also obtained from Baker Perkins.

2.2.2 PDX monomer puriﬁcation

All parts of the distillation setup were dried overnight at 110°C in a ventilated oven

before use. The distillation setup was then put together, purged by three vacuum/nitrogen

50

cycles and ﬁlled by a nitrogen ﬂow. The different parts of the distillation setup were held
by rubber bands attached via metal wires around the necks. Crude PDX monomer,
provided by Shell Chemicals and made by the catalytic dehydrogenation of ethylene
glycol using a copper (I/O) catalyst supported on silica particles is a mixture of PDX
monomer and linear or cyclic oligomers of PPDX (Shell Oil Company, US Patent).
Among the various puriﬁcation methods available, fractional distillation under reduced
pressure has been proven to be the most effective way to obtain pure PDX monomer with
high yield (98).

In practice, about 600 g of crude monomer was introduced into a previously oven-

dried and nitrogen-purged distillation system under nitrogen ﬂow (Refer Figure 2-13).

Three necked round bottom ﬂask
(crude PDX monomer)

Bﬁdge

Three necked round bottom ﬂask
(distilled PDX monomer)

   
  
   
 

Glass cap
/

H Vacuum
w l /‘
' ‘0

Heating mantle . \
Nitrogen

Magnetic stirrer

 

 

Ic wet bath
e or Three way valve

Figure 2-13: Schematic of Distillation setup

51

The distilled PDX monomer was recovered like a solid trapped by an ice/water bath (at
0°C). Starting from an initial purity degree of crude monomer of 90% such as determined
by 'H NMR in CDC13 , a recovery yield of 95% was obtained.

The residual acid content of the distilled monomer was determined by titration.
The titration consists in an acid-base reaction between the PDX monomer and KOH
using phenolphthalein as indicator. The medium is methanol at room temperature. An

average acid value of 2.3 mg K0,, / gPDx characterized the distilled monomer, which is
higher than the value required (below 0.5 mg KOH/ gpox) for typical ring-polymerization

of lactones such as CL (40) by reactive extrusion.
In order to reduce the acid value, the PDX monomer was again distilled. The acid
value obtained after the second distillation step decreased strongly from 2.3 to 0.33

mg K0,, / g PDX allowing us to envision the reactive extrusion polymerization of PDX

monomer.

It is very important to reduce the residual water content in the PDX monomer
before reactive extrusion. Water present could deactivate the active species in the
polymerization by the hydrolysis of the aluminum-alkoxide bond. Thus drying is a very
crucial step in the preparation of the PDX monomer for reactive extrusion. All glass parts
of the drying setup were dried overnight at 110°C in a ventilated oven before use. The
drying setup was then put together, purged by three vacuum/nitrogen cycles and ﬁlled by
nitrogen. The different parts of the drying setup were held by rubber bands attached via
metal wires around necks. The monomer (previously melted above its melting

temperature) was pumped through a molecular sieves (3A) column using a peristaltic

52

 

pump for at least 72 hr (Refer Figure 2-14). The PDX monomer, after distillation and

drying was polymerized by reactive extrusion.

 

 

 

Washing ﬂask
head 324/40 Peristaltic pump Red rubber hose
Nitrogen/vacuum >\ /
'\ F V Connection
B24/40
Molecular sieves
Two-way (3A) column
valve .
Filter (0.45
microns)
Three-necked 2L round

 

 

Bottomedﬂask (distilled ‘
PDX monomer)

Figure 2-14: Schematic of Drying setup

2.2.3 Batch Polymerization Studies

Before conducting extrusion experiments, the polyrnerizability of PDX and CL
monomers were determined in small scale by conducting polymerization experiments in
vials. This is also useful in determining the efﬁciency of the distillation and drying steps.
The initiator used for the polymerization is ATSB, which does not require any
puriﬁcation prior to the experiment. The initiator solution in toluene was prepared in a

glove bag under an inert atmosphere of nitrogen. 13 ml of anhydrous toluene was mixed

53

with 1.5 ml (to form a 10% solution) of ATSB. This solution was transferred to an 8 ml
vial, again under nitrogen ﬂow. A 10 ml conical bottom thin-walled vial was oven dried
and purged with nitrogen prior to the experiment. It was ﬁlled with 7 ml monomer and
sealed using a screw cap. The vials were heated using an oil bath at 140°C and 160°C for
the PDX and CL polymerization, respectively. For all the vials, nitrogen was supplied by
connecting the nitrogen hose to a syringe needle. Figure 2-15 shows the pictorial

representation of the experimental set up.

Nitrogen

 

> Syringe needle

I |———§ Conical bottom

 

 

 

 

—> Oil bath

 

 

 

 

 

Figure 2-15: Schematic of Batch polymerization setup

The theoretical degree of polymerization DP for the three-arm polymer is given by

equation (2.29) where X is monomer conversion and [M]0m, and[I]o‘m, signiﬁes the

initial moles of monomer and initiator charged to the reactor respectively.

54

_. [M 1...... X (2.29)

three-arm [1 law,

 

Thus the theoretical degree of polymerization of each linear arm of the three-arm
polymer is given by equation (2.30)

D _ D Rivet-arm (2,30)

Plincar - 3

Following this, Mu of the three-arm or the linear PCL polymer is given by equation
(2.31), where ‘114’ refers to the molar mass of CL monomer and M,l of the three-arm or
the linear PPDX polymer is given by equation (2.32), where ‘102’ refers to the molar

mass of PDX monomer.

M

n .three-arm / linear

= l 14(Dthee-ann/Iinear )g ["101 (231)

M = 102(Dthee-ann ”them )8 I mat (2°32)

n .three—arm / linear

The amount of initiator to be added was dependent on the desired molecular weight of the
polymer and was determined from equations (2.33) and (2.34) for PCL and PPDX

polymers respectively, where M OM, is the initial volume of monomer charged to the

reactor and p is the density of the monomer.

 

 

 

 

 

 

[I] = [M ]0.mol = [M ]0,mol = _ (,DMOM [114) = pMO'mI (2'33)
O’MOI Dl’rhree-aml (3M nJinear I 114) (3M n,linear / l 14) 3MnJin¢ar

[I] = [M 10M = [M]0.mol = (,oMo‘m, / 102) ___ 10M and (2.34)
03m”! D Rives-arm {3 M n,lr'near I 102) (3M nJinear [102) 3Mn,linear

Using the molar mass of aluminum tri-sec butoxide (97% purity) as 246 g/mol and the
density of the 10% (w/w) initiator solution in toluene as 0.874 g/ml, equations (2.33) and

(2.34) can be simpliﬁed further to equation (2.35).

55

 

0 mL - _
' (0.97)(O.874 g / mL)(0.10) M

n,linear

Equation (2.35) was used to calculate the necessary amount of initiator required to
synthesize a three-arm PCL and PPDX polymer with a desired Mn, um, (the number
average molecular weight of each arm in the three-arm polymer). The required amount of
initiator was withdrawn into a plastic syringe ﬁtted with a 23-gage needle in the glove
box. The tip of the syringe needle was capped with a rubber septum, and it was then
removed out of the glove box in order to be added to the preheated monomer. A
stopwatch was started when the initiator was added to the monomer in the reaction vial.
The reaction was allowed to proceed for the desired time, following which the vial was
removed from the hot bath and placed in liquid nitrogen to stop further reaction. Upon
cooling, the contents of the vial were removed for analysis of percent conversion of

monomer. The vial was broken to remove the product for analysis in the case of high

polymer.

2.2.4 Determination of Extent of Monomer Conversion

The extent of polymerization Or the monomer conversion to polymer was determined by
a dissolution-precipitation gravimetric method. A known amount, we, of the crude
polymer (around 5-10 grams) was dissolved in 50 ml solvent (anhydrous toluene in the
case of CL polymerization and 1,1,2,2-tetrachloroethane in the case of PDX
polymerization) for 24 hours at room temperature. The solution was then poured into a

ﬂask containing about 500 ml of another solvent (heptane in the case of CL

56

polymerization and Petroleum ether (Petroleum distillates, 30°C-60°C) in the case of
PDX polymerization) to precipitate the polymer and keep the un-reacted monomer in
solution. The precipitate was ﬁltered and dried in vacuum oven and weighed, wp. By
comparison to the starting crude polymer weight, the fractional monomer conversion, X

was calculated as explained in equation (2.36).

X = 2:”. (2-36)

2.2.5 Extrusion Polymerization Studies

The molecular architecture of the PPDX polymer produced by the ROP of PDX

monomer initiated by ATSB is as shown in Figure 2-16.

 

O
... + ;
O.

Monomer Catalyst Three-arm
polymer

Figure 2-16: Three-arm structure of the PPDX polymer produced by reactive

extrusion polymerization of PDX monomer.

57

Due to the expected three-arm nature of the polymer produced, the desired theoretical Mn
of the PPDX polymer is designated as the molecular weight of each individual arm of the
three-arm polymer. The relative ratio of the monomer to the initiator determines
molecular weight of the polymer produced. Due to limited availability of the PDX
monomer and the difﬁculties associated with repeated distillation and drying operations,
the optimal conditions for extrusion (temperature, screw speed and screw conﬁguration)
were determined by conducting the polymerization of CL monomer. Table 2-5 shows the

processing parameters along with the desired theoretical molecular weights.

Table 2-5: Extrusion Processing Conditions for CL polymerization at 150°C.

 

 

 

 

RPM M M R“) I S R b) Theoretical. Torque
(g/ min) (g/min) MnPCL (%)
(each arm,
g/mol) c
130 96 0.41 193,780 61
130 328 0.61 44,400 60
130 482 0.61 65,300 57

 

 

 

 

 

 

 

a) Mass ﬂow rate of monomer
b) Mass ﬂow rate of initiator solution

c) Theoretical Number average molecular weight (g/mol) of each arm determined

from Equation (2.32)

58

Knowing the polymerizability of CL monomer to three different molecular weights, the
synthesis of REX-PPDX by extrusion polymerization of PDX monomer using ATSB
initiator in a twin-screw co-rotating extruder was conducted. Speciﬁcally, the
polymerization was conducted at three different temperatures for three molecular
weights. The monomer and initiator ﬂow rates required for a given desired theoretical
molecular weight was calculated as shown from equation (2.29) through equation (2.35).
A peristaltic pump with MASTERFLEX tubing (Pharrned 06485-15) was used to meter
the PDX monomer to the extruder feed zone. The PDX monomer was heated using a
heating mantle and the temperature of the monomer maintained at around 35°C to enable
facile pumping of the monomer into the extruder. Nishida et al have explained the density
of the monomer as a function of temperature (37). The initiator solution was pumped
using a MASTERFLEX pump with Master ﬂex tubing (06412-14). The extrusion
processing conditions for the synthesis of REX-PPDX is shown in Table 2-6. The overall

process schematic is shown in Figure 2-17.

Nitrogen

Nitrogen Nitrogen

     

Nitrogen

 

 

‘m-r- L41~~.JL .TL. ' safari-:4" r-r .
r, set." mama. .; r W .di D Die

 

Figure 2-17: Process schematic for REX-PPDX production

59

The pumps were each calibrated over the desired flow rate range in order to accurately

control the process by adjusting the ratio of the monomer to the initiator. (Refer Figure

2-18 and Figure 2-19). The PPDX homo polymer was collected in aluminum pans and

   

 

 

air-cooled.
0.8 -
0.6 -
D
C
'5 0 4 -
.3 y = 00065:: - 0.0069
0-2 ' R2 = 0.9999
0 . . .
0 50 100 150

Flow rate (glmin)

Figure 2-18: PDX monomer pump calibration curve (PDX monomer maintained

between 32-35%)

   

 

 

0.8 -
0.6 -
D
C
'5 0.4 .
.3 y = 0.0065x - 0.0069
0-2 ' R2 = 0.9999
0 I I I
0 50 100 150

Flow rate (glmin)

Figure 2-19: ATSB initiator solution calibration curve

60

Table 2-6: Polymerization of PDX monomer by Reactive Extrusion: Determination

 

 

 

 

 

 

 

 

 

of Optimal Conditions.
Tm], M M R“) I s R b) Theoretical Torque
(°C) (g/min) (g/min) Mn PPDX (%)
(each arm,
g/mol)
130 23.23 0.60 63,400 14
130 23.23 0.41 94,550 14
130 40.75 1.22 54,800 14
150 23.23 0.60 63,400 14
150 23.23 0.41 94,550 . 14
150 40.75 1.22 54,800 14
170 23.23 0.60 63,400 14
170 23.23 0.41 54,800 14

 

 

 

 

 

 

 

a) Monomer Mass Rate

b) Initiator Solution Mass Rate

In contrast to strong strands of PCL, PPDX strands are very low viscosity and liquid like.
These values can be related to the low values of torque as observed and explained in

Table 2-6.

61

PDX monomer was co-polymerized with three different mole fractions of CL monomer
to produce reactively extruded PCL-co-PPDX polymers. The feed set up was very similar
to the case of homo polymerization of PDX monomer with the CL monomer being
pumped to the feed port of the extruder. The Degree of Polymerization of the copolymer

can be obtained using equation (2.37) as follows.

 

DP _ [M ]0,molCL + [M ]O,moIPDX (2,37)
CL—co-PDX - [110 M

The theoretical desired molecular weight of the co-polymer can be calculated knowing
the relative rates of the PDX monomer, CL monomer and the initiator solutions. The

extrusion processing conditions are shown in Table 2-7.

62

Table 2-7: PCL-co-PDX by reactive extrusion at 130°C: Effect of the addition of

limited amounts of CL monomer

 

 

 

 

 

 

 

 

 

 

 

 

 

PM R” I s R 5’ c M R “7 Theoretical Imposed Torque
(g/min) (g/min) (g/min) Mn PPDX ‘ For. (%)
(each arm,
g/mol)
23.23 0.62 - 94,550 - 14
23.23 0.407 2.369 104,200 0.08 9
40.7 1.21 3.294 107,950 0.11 16
23.23 0.60 5.12 77,550 0.16 19
a) PDX Mass Rate

b) Initiator Solution Mass Rate

c) CL Mass rate

2.2.5.1 Screw Configuration

The two types of elements that make up the screw in a twin-screw extruder are conveying
and kneading elements as discussed earlier. Krishnan (40) has shown that use of a
conveying + kneading screw conﬁguration leads to extra hold up time in the kneading

blocks, sufﬁcient to increase the viscosity of the polymer in that section to a high level.

63

On account of this, the polymer-monomer mixture preceding that section cannot be
conveyed forward against the backpressure generated by the “frozen” high molecular
weight polymer melt. This could result in complete stoppage of flow of material through
the extruder. However, when an entirely conveying screw conﬁguration was employed,
there were no problems associated with production of high molecular weight polymer.
This study employed a completely conveying screw conﬁguration. The screw
conﬁguration is shown in Figure 2-20. All polymerization experiments were carried out
using the same screw conﬁguration under different conditions of temperature, desired
molecular weight, screw speed and throughput. Various extruder temperatures studied
ranged from 130 to 180°C; the monomer to initiator feed ratios was adjusted to give the
desired theoretical number-average molecular weight, M..; the monomer feed rate, which
is the throughput of the extrudate was set at two values of 23.23 and 40.75 gmlmin. A
screw speed of 130 RPM was used for the study. The percent conversion of the monomer
was obtained for each condition. RTD studies were also conducted using the same screw

conﬁguration.

 

Extruded Polymer melt

 

Figure 2-20: Screw Configuration evaluated for the Synthesis of REX-PPDX

2.2.5.2 Residence Time Distribution Studies

Residence time distribution studies were conducted for the polymerization of CL
monomer in a BAKER PERKIN twin-screw extruder using an instantaneous charge of 50
mg of a ﬂuorescent die, UV-XPBB, obtained from LDP NET. The die has a fluorescence
emission and excitation at 453 nm and 365 nm respectively. RTD studies were conducted
for the various operating conditions as shown in Table 2-8. M,l of the resulting PCL
polymer was kept constant at 70,000 g/mol. Extrudate samples were collected every 30
seconds for the ﬁrst three minutes and further on, every one minute up to 2 minutes. The
samples were then dissolved in toluene and the concentration of the dye obtained using
UV-Spectrosc0py. The result was then plotted against time to give the distribution, E (t).
The cumulative distribution, F (0) was also plotted to determine the applicability of the
Axial Dispersion model. Peclet no (Pe,) values were ﬁt to a model deemed appropriate,
using STATEASE ‘Design of Experiment’ 6.0 modeling software. Conversion of the
monomer to polymer was determined for all of the above cases (Section 2.2.2).
Experimentally observed conversion values were compared to conversion values

obtained from the model shown in equation (2.22).

65

Table 2-8: Experimental Scheme to synthesize REX-PCL for RTD studies.

 

 

 

 

 

 

 

 

 

 

 

 

Desired Throughput (ml/min) Extrusion Temperature ("C) Screw Speed (RPM)
14 130 125
22 130 125
14 130 225
22 130 225
14 170 125
22 170 125
14 170 225
22 170 225

 

2.2.6 Intrinsic Viscosity

Intrinsic viscosity [77] reﬂects the contribution of the polymeric solute to the difference

between the viscosity of the mixture and that of the solvent. Intrinsic viscosity is
determined using an Ubbelohde viscometer. The viscometers are available in different
sizes, and the type and size to be used depends on the range of viscosity of a particular
polymer/solvent combination. In this study, size 0C was used. Firstly, the dilute polymer
solution of known concentration was made by dissolving pre-weighed polymer (between
0.1 to 1.5 g depending on the type of polymer) into a known volume of solvent, usually

between 50 to 100 ml. For the case of PCL polymer, toluene was used as the solvent

66

 

while a 3:2 v/v mixture of tetrachloroethane/phenol solvent used for the PPDX polymer.
Typically efﬂux times (to be discussed later) are between 5-10 minutes for the particular
concentration of the polymer. Both the viscometer and the solution to be tested were
made particle-free to prevent clogging of the capillary tube inside the viscometer.

The viscometer was cleaned by ﬂushing solvent through all the three tubes,
followed by dry air to remove ﬁnal traces of solvent. The polymer solution to be tested
was then introduced through the largest tube (tube A) till the level of the solution was
between the two lines at the bottom of the tube. The viscometer was then placed in a
constant-temperature water bath (temperature maintained at 24°C) such that the water
level was above the bulb on tube B. The viscometer was then left for at least 20 rrrinutes
to bring the water and the solution temperature to be at equilibrium. The Ubbelohde was
vertically aligned inside the water bath, a ﬁnger was placed over the third tube (tube C)
and suction applied to tube B, until the solution reached to the bulb on the tube B.
Suction was then removed from tube B, followed by removing the ﬁnger from tube C and
the solution was allowed to ﬂow freely. The efﬂux time was determined by measuring
the time required for the level of solution in tube B to drop from one marking to the
other. The above procedure was repeated at least three times for each solution and the
average efﬂux time was determined.

Either the Huggins equation or the Kraemer equation can be used to determine
intrinsic viscosity. These equations are explained in (2.38) [Huggins] and (2.39)

[Kraemer] below.

67

c"(77/ 770 -l)= [771+ kn [77126 (2.38)

(34 ln(ﬂ/ 770 ) = [77]_ k1 [7712 c (2°39)

where c is the concentration of the polymer solution tested, 7] is the solution viscosity,
770 is the solvent viscosity, [7}] is the intrinsic viscosity and k” and k1 are the

proportional constants associated with each equation. Both constants depend on the type

of polymers being tested. For dilute solutions, it can be said that:

77/ 770 = t/ to (2.40)

where t is the efﬂux time of dilute polymer solution and to is the efﬂux time of solvent
alone. The intrinsic viscosity was determined as the y-intercept obtained by plotting the
left—hand-side of either the Huggins or the Kraemer equation against concentration.

Mark-Houwink-Sakurada (MHS) equation relates intrinsic viscosity ([77]) of a

polymer to the viscosity average molecular weight (M v)with the equation (2.41)
[77] = KM 5’ (2.41)

MHS constants, ‘K ' and ‘a’ depend on the polymer, type of solvent used and the solution
temperature used for testing. Since MHS constants, representing a wide range of
polymers with different types of solvents and testing temperatures, have been published
in various literatures, M, can be determined if the intrinsic viscosity of a polymer is
determined at the condition in which MHS constants are known. If MHS constants are

unknown for a particular combination of polymer, solvent, and testing temperature, they

can be calibrated by determining [77] and plotting ln[77]versus ln(M w)or ln(M“ ). ‘K’ and

‘a’ are determined as the slope and intercept of the above plot. This method, however, is

68

merely an estimate and determination of exact My involves fractionation of the whole
polymer sample into subspecies of relatively narrow molecular weight distributions and
then averaging the measured Mn and MW to calculate My.

In general, MHS equation can be applied to estimate MW or Mn of unknown
samples, only if a fraction of a molecular weight distribution is very similar to the sample
used for the calibration. However, this restriction does not apply to the polymers with
random molecular weight distribution, such as linear polyarrrides and polyesters

polymerized under equilibrium conditions. For these types of polymers, MW = 2M" and

the relation [77] = KM f can be applied.

2.2.7 Thermal Analysis

Standard DSC measurements were carried out with a 2920 CE Modulated DSC from TA
Instruments. Samples, after extraction of the aluminum alkoxide functions by
liquid/liquid extraction, were heated from -1 10°C to 140°C at the rate of 10°C per minute
under nitrogen ﬂow. Glass transition temperature (T g) and Melting temperature (T m) were
determined from the plot.

TGA measurements were performed, by using a Hi-Res TGA 2950 of TA
Instruments under helium ﬂow. Samples were heated from room temperature to 180°C at
the rate of 20°C per minute and held at 180°C for 70 minutes. The thermal stability of the

various homo and copolymers were determined from the plot.

69

2.2.8 UV-Spectroscopy

UV/V is Spectroscophotometer ‘Lambda 25’ from Perkin Elmer was used to determine
the concentration of the ﬂuorescent dye in the extruded PCL sample. This concentration
data was further used to determine the RTD at a given temperature, screw speed and
overall throughput. PCL sample, containing the dye, was dissolved in toluene.
Concentrations of the solution were in the range of 10'3 to 10'5 M. Samples were put into
glass cuvettes (plastic cuvettes dissolved in toluene). A reference containing the same
concentration of the PCL sample in toluene (without the dye) was used. A spectrum of
Absorbance vs. Wavelength was obtained. Only absorbance values in the range of 0.15
to 1.5 are valid. A peak was obtained corresponding to 370 nm,’which is the ﬂuorescence
excitation of UV-XPBB. Am, the wavelength at which the molecule absorbs the most
light was obtained as the maximum value of the peak. A calibration curve of Am vs.
Concentration was obtained by dissolving known amounts of the dye in the PCL solution.
(The concentration of PCL in toluene was kept constant). Using Beer’s law, which states
that the number of photons absorbed is proportional to the concentration of the substance
in solution, the actual concentration values were interpolated from the calibration data.
Care must be taken that the solution does not absorb any light in the same range as that of

the solute. Figure 2-21 shows a block diagram of UV-Spectroscopy.

 

 

 

 

 

 

 

 

---\--- reference -- detector
mono- -

lamp chrom ator ratio

\‘n- sample r-udetector—r

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 2-21: UV Set Up Block Diagram

70

2.3 RESULTS AND DISCUSSION

The results of batch polymerization studies of CL and PDX monomer using ATSB
initiator will be explained. The results of the effects of different temperatures, screw
speeds and throughput on monomer conversion and molecular weight in the extrusion of
PPDX polymer will be discussed in this section. Moreover, the effect of addition of
different mole percentages of CL monomer on monomer conversion during the bulk
extrusion polymerization of PDX will be discussed. Thermal analysis results of PPDX
and PCL-co-PPDX polymers will be presented and discussed. The results of RTD studies
on the polymerization of CL monomer will also be presented. The suitability of the Axial
Dispersed PFR model on RTD and CL conversion in the extrusion polymerization of CL

monomer will be evaluated.

2.3.1 Batch Polymerization Studies

The conversion results of the batch polymerization experiments of CL and PDX
monomers are shown in Table 2-9. It can be observed that the conversion of CL
monomer is greater than 95% as opposed to a 71% conversion of the PDX monomer. As
known, since the PDX polymerization is characterized by monomer-polymer equilibrium
above the melting temperature of the polymer, conversion values of around 71% is
permissible. These experiments further lead us to test the polymerizability in a twin-

screw extruder.

71

Table 2-9: Conversions 0f PDX and CL monomers in experimental conditions close

to Reactive Extrusion (Polymerization time = 2 minutes)

 

 

 

Experiment Lactone Tpoly Conversion
(°C)
1 PDX 140 71
2 CL 160 96

 

 

 

 

 

 

2.3.2 Effect of extrusion process parameters on CL and PDX monomer conversion

For the various extrusion process conditions shown in Table 2-5, the conversion of CL
monomer was 99%. Torque values remained constant (60%) even as the molecular
weight increased from 44,400 g/mol to 193,780 g/mol. Thus, it is clear that at a
temperature of 150°C, for different overall throughputs, using a completely conveying
screw conﬁguration, near complete conversion is observed (99). PCL polymer strands
were of high viscosity and could be easily pelletized by quenching in a water bath.

The conversion results of PDX homopolymerization (Processing conditions

shown in Table 2-6) by reactive extrusion are shown in Table 2-10.

72

Table 2-10: Polymerization of PDX monomer by Reactive Extrusion: Determination

of optimal Conditions.

 

 

 

 

 

 

 

 

 

Tpoly M M R“) I S RE Theoretical Torque Conversion c)
(°C) (g/ min) (g/min) Mn PPDX (%) (%)
(each arm,
g/mol)

130 23.23 0.60 63,400 14 68
130 23.23 0.41 94,550 14 78
130 40.75 1.22 54,800 14 68
150 23.23 0.60 63,400 14 62
150 23.23 0.41 94,550 14 ‘ 68
150 40.75 1.22 54,800 14 60
170 23.23 0.60 63,400 14 61
170 23.23 0.41 54,800 14 6O

 

 

 

 

 

 

 

 

a) Monomer Mass Rate
b) Initiator Solution Mass Rate

c) Determined by gravimetry

In contrast to strong strands of PCL, the strands of PPDX are low viscosity. From Table
2-10, based on the conversion values; it is evident that the homo polymerization of PDX
proceeds similarly to that of a system using a standard vacuum technique. For example,
the monomer conversion at 130°C is 68% and 66% for the polymerization of PDX by
reactive extrusion and standard vacuum technique respectively. In the case of the homo

polymerization of PDX monomer to PPDX polymer, irrespective of the monomer ﬂow

73

rate used, the maximum conversion obtainable was limited to about 78% which is close
to the thermodynamic equilibrium monomer conversion value. It is also observed that for
quite identical theoretical molecular weights (54, 800 g/mol and 63,400 g/mol),
conversions are identical even as monomer ﬂow rate increases from 23.23 to 40.75
gm/min. It is evident that as temperature increases, monomer conversion decreases due to
the dynamic chemical equilibrium between the monomer and the polymer. It is also
evident from Table 2-10 that as the monomer ﬂow rate increases from 23 to 40 gmlrnin,
the conversion drops. This is in agreement with batch polymerization studies done by
various researchers (24,37,38 and 39) who report that the conversion of PDX monomer
passes through maximum and then levels off at the equilibrium monomer conversion

value.

2.3.3 Effect of CL addition on PDX Polymerization

The conversion results of PDX monomer on addition of a few—mol% CL during the PDX

extrusion polymerization are shown in Table 2-11.

74

Table 2-11: Polymerization of PDX by reactive extrusion: Effect of the addition of

limited amounts of CL monomer

 

 

 

 

 

 

 

 

 

 

 

 

 

P M R“) I S R b) C M R C) Theoretical Imposed Torque Conv. d)
(g/min) (g/min) (g/min) MnPPDX fad) (%) (%)
(each arm,
(g/mol)
23.23 0.60 - 94,550 - 14 65
23.23 0.41 2.37 104,200 0.08 19 99
40.75 1.22 3.29 107,950 0.11 16 99
23.23 0.60 5.18 77,550 0.16 19 100
a) PDX Mass Rate

b) Initiator Solution Mass Rate
c) CL Mass rate

d) Determined by gravimetry

It is clear from Table 2-11 that with incorporation of 8%, 11% and 16% CL monomer,
the conversion is almost complete (~ 100%). This makes the process economical leading

to the formation of a melt-stable blocky copolymer structure.

75

 

2.3.4 Intrinsic Viscosity Results

The results of the intrinsic viscosity experiments conducted on PPDX and PCL-co-PPDX
polymers are shown in Table 2-12. It can be noted from the Table 2-12 that there is a
sharp decrease in the [7)] values on incorporation of CL units into the PPDX chain.
However, the molecular weight of the polymers remains constant (ﬁrst three entries).
Thus, the decrease in [77] values could be attributed to a change in the conformation of
the copolymers in solution. Unfortunately, it was not possible to determine the actual
molecular weight of the copolymers by Size Exclusion Chromatography (SEC) or
osmometry due to the poor solubility of the obtained copolymers in conventional organic

solvents.

76

Table 2-12: Intrinsic viscosity of PPDX and PCL-co-PPDX polymers produced by

reactive extrusion (T=130”C and Screw Speed = 130 RPM)

 

 

 

 

 

P M R“) IS R“ c M R“) Imp. FCL Conv d’ [5]”
(g/min) (g/min) (g/min) (%) (deg)
23.23 0.60 - - 65 0.77
23.23 0.41 2.37 0.08 99 0.65
40.75 1.22 3.29 0.11 99 0.55
23.23 0.60 5.18 0.16 100 0.37

 

 

 

 

 

 

 

 

a) PDX mass ﬂow rate

b) Initiator solution mass ﬂow rate

c) CL mass ﬂow rate

d) As determined by gravimetry after solubilization of crude product in l, l, 2, 2-
tetrachloroethane, precipitation in 7 volumes of heptane, ﬁltration and drying up to
constant weight.

e) As determined by solution viscometry in l, l, 2, 2-tetrachloroethanelphenol (3:2 v/v)

at 25°C.

77

2.3.5 Thermal Analysis Results

DSC therrnograrns of PPDX and PCL-co-PPDX polymers produced by reactive extrusion
are shown in Figure 2-22. The effect of CL addition on the thermal transitions is

explained in Table 2-13.

L .L A

Heat Flow (Wig)

cameos.

-1. x:
-2 _ 5

V v ' V

.110 -60 -10 40 90 140
Exo Up Temperature (C)

 

 

 

Figure 2-22: DSC thermogram of PPDX homopolymer and PCL-co-PPDX
copolymers obtained by reactive extrusion with different final molar fractions in
CL: FCL= 0.00 (a), 0.11 (b), 0.16 (c) and 0.20 (d) (heating rate = 10°C/min, under N2,

only the second run is considered)

As expected, the incorporation of around 11-mol% CL into PPDX chains does not
signiﬁcantly modify the thermal properties. The co-polymer is characterized by a melting

temperature around 94°C, close to the homopolymer melting temperature value of 106°C.

78

Moreover, there is an interesting internal plasticizing effect as the glass transition
temperature of the co-polymer decreases from -1l°C (homo PPDX) to -220C. The
melting temperature of the co-polymers further reduces to 85°C and 68°C with
incorporation of 16-mol% and 24-mol% CL respectively. The glass transition
temperatures also decrease with increasing amount of CL. As the molar fraction in CL in
the copolymers increases from 0 to 20- mol%, the melting enthalpy and melting
temperature decrease smoothly. For FCL = 0.20, the melting dip centered at 68°C is very
broad and the ability of the copolymer to crystallize is highly limited as evidenced by the
low value of crystallization enthalpy. This low melting temperature of the copolymer

limits the available service temperature range.

Table 2-13: DSC data of PPDX and PCL-co-PPDX polymers as obtained by reactive

extrusion (under nitrogen flow, from -100°C to 140°C at a heating rate of 10°Clmin,

 

 

 

 

 

 

2""l scan)
Entry Imposed Tg AHc Tc AHm " Tm
FCL (°C) (J/g) (°C) (1, g) (°C)
1 0 -1 1 38 22 69 106
2 0.1 1 -22 33 13 56 94
3 0.16 -23 36 19 48 85
4 0.20 ~29 27 29 32 68

 

 

 

 

 

 

 

79

 

 

 

105 ~
1 00"

95‘ (d)
. 0 .

Welght ( A) ‘ -.. (c)
(b)
85 . . . . . a .
0 1 0 20 30 40 50 60 70

Time (min)

Figure 2-23: Time-dependence of weight loss at 180°C for PPDX and PCL-co-PPDX
polymers obtained by reactive extrusion with different final molar fractions in CL:

FCL= 0.00 (a), 0.11 (b), 0.16 (c) and 0.20 (d) (T = 180°C under a He flew)

Figure 2-23 shows the isothermal therrnogravimetric analyses of the polymers obtained
by reactive extrusion. As previously mentioned for DSC measurements, aluminum
residues were extracted by liquid-liquid extraction. It can be seen from Figure 2-23 that
the thermal stability of the copolymers is substantially improved by the incorporation of
CL into PPDX chains. The rate of weight loss of the PPDX homopolymer and the PCL-
co-PPDX polymers are shown in Figure 2-24. This is in perfect agreement with our
previous observations and the formation of a blocky-like copolymer structure. Moreover,
this enhancement of the thermal stability is much higher than the one achieved by a

simple end-capping process (Shell Process)

80

0.2 -
0.184
0.16‘
0.14 ‘

0.1 '
0.08'

0.06 -
0.04 1

0.02 ‘
. .

-(dWIdT)%/min

 

 

a b c d
PPDX and co-polymers with PCL

Figure 2-24: Rate of weight loss for PPDX and PCL-co-PPDX polymers obtained by
reactive extrusion with different final molar fractions in CL: FCL= 0.00 (a), 0.11 (b),

0.16 (c) and 0.20 (d) (T = 180°C under a He flow)

2.3.6 Residence Time Distribution Studies

RTD studies were conducted for the reactive extrusion polymerization of CL monomer
using the conditions mentioned in Table 2-8. A completely conveying screw
conﬁguration was employed for the polymerization process. Figure 2-25 and Figure 2-26
show the RTD data. It can be observed that at both 130°C and 170°C, for a constant
monomer feed rate, increasing the screw speed from 125 to 225 RPM slightly decreases

the peak value of residence time. However, the major reduction in the peak residence

81

time value is obtained when the monomer feed rate is increased from 14 to 22 mllmin at a
constant screw speed and temperature. In all the cases, a broad distribution of residence
times was observed, with a fair amount of tailing, despite the excellent self-wiping
characteristic of a co-rotating twin-screw extruder. This can be attributed to the presence
of liquid ﬁlms of monomer which would be present in the screw elements and will be
eluted only after very long times. The RTD data at 170°C is very similar to the RTD data
at 130°C for the same screw speed and monomer feed rate. This is expected since the

only inﬂuence of the higher temperature was to accelerate the polymerization kinetics.

 

 

   
   
  
  

 

 

   

 

 

 

 

+ 130C, 125 RPM, 14
0.4 i .
mllmin monomer
0.35 . me
I 130C, 125 RPM, 22
0.3 - mllmin monomer
rate
0.25 - - v 130C, 225 RPM, 14
mllmin monomer
a 0.2 " rate
E 0 15 q ; —x— 130C, 225 RPM, 22
' mllmin monomer
0.1 _ rate
0.05 -
0 .;.
20
-0.05

Time (min)

Figure 2.25: RTD curves at 130°C

82

 

 

 
 
  
 
  

 

 

 

 

 

 

0.35 - --+- 170C, 125 RPM, 14
. mllmin monomer rate
0-3 ' -‘ — a. -170C, 125 RPM, 22
0 25 mllmin monomer rate
’ -—ev - 170C, 225 RPM, 14
0.2 .. mllmin monomer rate
A + 170C, 225 RPM, 22
S 0.15 . mllmin monomer rate
0.1 -
0.05 a
0
20
-0.05

Time (min)

Figure 2-26: RTD curves at 170°C

2.3.6.1 Reactive Extrusion Modeling

The above RTD data for polymerization of CL monomer in a co-rotating twin-screw
extruder was analyzed to determine if it ﬁt the Axial Dispersed Plug Flow Reactor Model
discussed earlier. Figure 227 through Figure 2-34 represent the prediction of RTD by the
Axial Dispersed Plug Flow model [Refer equations (2.16),(2.17) and (2.18)]. From the
ﬁgures, it can be seen that the Axial Dispersed PFR model is a suitable choice to predict
ﬂow in a co-rotating extruder for the case of CL polymerization. A plot of FM.) vs.

Emu-mm; yielded ﬁt coefﬁcients ‘b’ (Fma / chpumm, ) in the range of 0.92 to 0.98.
Table 2-14 presents the ﬁt coefﬁcients for the various conditions.

83

Table 2-14: Fit coefficients ‘b’ predicting goodness of fit of the Axial Dispersed PFR

model to experimental data in the case of CL polymerization

 

 

 

 

 

 

 

 

 

 

Desired Throughput (ml/min) Extrusion Temperature (°C) Screw Speed (RPM) B
14 130 125 0.94
22 130 125 0.90
14 130 225 0.94
22 130 225 0.98
14 170 125 0.96
22 170 125 0.96
14 170 225 0.96
22 170 225 0.97

 

 

 

 

84

 

    

 

+F experimental

 

 

 

+F model
tr. 0.6 d *
0.4 J
0.2 4
0 u u v
0 5 10 15 20

time (minutes)

Figure 2-27: Axial Dispersed PFR model applied to RTD data for CL

polymerization at 130°C (Screw speed of 125 RPM and monomer feed rate of 14

mllmin)

 
   

 

0'8 " +F experimental

h 0.6 . +F model

 

0.4 ..

0.2-1

 

 

0 5 10 15 20

time (min)

Figure 2-28: Axial Dispersed PFR model applied to RTD data for CL
polymerization at 130°C (Screw speed of 125 RPM and monomer feed rate of 22
mllmin)

85

   
  

 

+ F experimental
-I— F model

 

 

 

 

I U I

0 5 10 15

time (minute)

Figure 2-29: Axial Dispersed PFR model applied to RTD data for CL

polymerization at 130°C (Screw speed of 225 RPM and monomer feed rate of 14

mllmin)

 

 

 

 

Figure 2-30: Axial Dispersed PFR model applied to RTD data for CL
polymerization at 130°C (Screw speed of 225 RPM and monomer feed rate of 22

mllmin)

86

1.2-

     

11

 

+ F experimental ’
-I- F model

0.8 .

{=- 0.6'

0.4 -

0.2 -

 

 

U U I U

0 5 10 15 20
time (min)

Figure 2-31: Axial Dispersed PFR model applied to RTD data for CL
polymerization at 170°C (Screw speed of 125 RPM and monomer feed rate of 14

mllmin)

  

 

1.2 -
1 e
03 ., +F experimental
+17 model

 

In 0.6-

0.4 -

0.2-l

 

 

0 2 4 6 8 10 12 14

Figure 2-32: Axial Dispersed PFR model applied to RTD data for CL
polymerization at 170°C (Screw speed of 125 RPM and monomer feed rate of 22
mllmin)

87

  

 

 

 

 

 

1 I
0 8 . +F experimental
-I- F model
In 0.6 «-
0.4 -
0.2 -
0 u 1
0 5 10 15 20
time (min)

Figure 2-33: Axial Dispersed PFR model applied to RTD data for CL

polymerization at 170°C (Screw speed of 225 RPM and monomer feed rate of 14

mllmin)

 

 

 

 

 

1.2 -
l I
+F experimental
0'8 ‘ +F model
in 0.6 -
0.4 d
0.2 a
0 I I I
0 5 10 15 20

time

Figure 2-34: Axial Dispersed PFR model applied to RTD data for CL
polymerization at 170°C (Screw speed of 225 RPM and monomer feed rate of 22
mllmin)

88

Peclet number (Pe,) was calculated for the various experimental conditions shown in

Table 2-8 using equation (2.16). The Peclet number values for the various experimental

conditions are shown in Table 2-15.

Table 2-15: Peclet Number values under different experimental processing

conditions (Table 2-8) in the synthesis of three-arm PCL polymer

 

 

 

 

 

 

 

 

 

 

Desired Throughput Extrusion Temperature Screw Speed Peclet
(ml/min) (°C) (RPM) No
14 130 125 6.41
22 130 125 6.8
14 130 225 3.03
22 130 225 9.98
14 170 125 1 1.56
22 170 125 5.26
14 170 225 9.31
22 170 225 7.49

 

 

 

 

It is observed from Table 2-15 that maximum dispersion is obtained under conditions of

130°C, 225 RPM and 14-ml/min monomer feed rate. Least dispersion is obtained at

170°C, 125 RPM and l4-ml/min monomer feed rate. At 130°C, due to the presence of

low molecular weight pre-polymers along-with the polymer melt, dispersion effects

become pronounced. Moreover, dispersion effects increase as overall throughput

increases (increase of monomer feed rate).

89

 

A 2F] model was ﬁt using STATEASE ‘Design of Experiment’ 6.0 modeling software
with Pe, as the response data and Screw Speed, temperature and thruput as the factors.
From the model, the signiﬁcant terms were determined to be temperature, interaction of
screw speed, and thruput and interaction of temperature and thruput. The ﬁnal model
equation in terms of the actual factors is shown in equation (2.42).

Fe, = —42.41 — 0.12 * Screwspeed + 0.48 * Temperature + 2.39 * Thruput (2.42)
+ 0.007 * Screwspeed * Temperature - 0.02 * Temperature * Thruput

The experimentally determined conversions were compared to the ones obtained from the
model. The ﬁt is shown in Figure 2-35. It can be observed from Figure 2-35 that the

model provides sort of a good ﬁt to experimental data.

90

Conversion (Model)

0.99 -
0.98 ‘

0.97 - O
. O

P P P
\e \e \e
A u: a
l l I

c e
{e ie
N u
l j

0

0.91 ‘

 

 

0.9 I U U U r I U U U

0.9 0.91 0.92 0.93 0.94 0.95 0.96 0.97 0.98 0.99

Conversion (Experimental)

Figure 2-35: Model vs. Experimental conversions in the polymerization of CL

monomer to form three-arm PCL polymer (Table 2-8).

91

1

Chapter 3

 

REACTIVE EXTRUSION CROSSLINKING OF BIODEGRADABLE

POLYESTER AND BLENDING WITH INORGANIC FILLERS.

 

3.1 INTRODUCTION

3.1.1 Cross-linking Chemistry

Cross-linking is typically carried out to increase molecular weights of polymers to make
them suitable for practical use. Ostromislenski ﬁrst reported the initial use of organic
peroxide as a cross-linking agent in 1915 for the vulcanization of natural rubber. Since
around 1950, the interest in the industrial use of peroxides as cross-linking agents has

been increasing, partly as a result of the introduction of saturated rubbers such as

92

ethylene-propylene monomer and silicone rubber which cannot be vulcanized with the
usual sulfur systems (100,101). On the other hand, many researchers have tried to
introduce cross-links into aliphatic polyesters to improve properties such as mechanical
strength and elasticity. Han et al. (102) prepared biomedical materials by the thermal
cross-linking of unsaturated polyesters using free radical initiators. Pramanick and Ray
synthesized cross-linked polyesters from a mixture of glycerol, citric acid and aspartic
acid (103,104). A schematic showing a cross-linked polymer network as opposed to

uncross- linked one is shown in the Figure 3-1 below.

“‘93 ”11:19
f

Uncross-Ilnlted Cross-linked
network network

 

 

 

 

 

 

 

 

 

Figure 3-1: Schematic showing cross-linked and uncross-linked networks

Depending on the type and the amount of initiator used, polyesters can be highly cross-
linked to form a therrnoset or lightly cross-linked to provide more elastomeric properties.
There are numerous patents in the area of cross-linking biodegradable polyesters to use as
binders in toner applications. US. Patent 3, 681,106 provide a method to modify a

polyester resin by mixing a trivalent or more polyol or polyacid with the monomer to

93

generate branching during polycondensation. In US. Patent 3, 941,898, a cross-linked
vinyl type polymer was prepared by conventional cross-linking. The patents also describe
that the initiator, whether solid or liquid, is mixed with the polyester resin, for example in
a rotary tumble blender or a batch mixer. US. Patent Application 20020086944 provides
a method to cross-link rubber particles using metal peroxides in extruders. US. Patent
6,359,105 provides a method to form cross-linked polyester resins using a liquid
chemical initiator such as 1, l-bis (t-butyl peroxy)-3, 3, 5-t1imethylcyclohexane. The
method includes spraying the initiator onto the unsaturated polyester resin prior to, during
or subsequent to melting of the unsaturated polyester resin to form a polymer melt and
subsequently cross-linking the polymer melt under high shear. In all of the above cases,

cross-linking is carried out to improve the xerographic performance of the toner.

3.1.2 Cross-linking Principles

Chemical cross-linking is one of the most severe mechanisms for decreasing molecular
freedom. It links the polymer chains together through covalent or ionic bonds to form a
network. Cross-linking can be achieved in two broad ways: 1) cross-linking during
polymerization by use of polyfunctional instead of difunctional monomers, 2) cross-
linking using external initiators such as peroxides. The cross-links may contain the same
structural features as the main chains, which are usually the case with the former, or they
may have an entirely different structure, which is more characteristic of the latter.

A number of changes occur due to cross-linking. If previously soluble in a

particular solvent, the cross-linked polymer will be partially or totally insoluble in the

94

same solvent. When the cross-linked polymer is dissolved in a good solvent, the polymer
absorbs a portion of the solvent and subsequently swells. The extent of swelling is a
competition between two forces. The free energy of mixing will cause the solvent to
penetrate and try to dilute the polymer solution. Increasing the temperature may enhance
this entropic increase. As the polymer chains in the cross-linked polymer network begin
to elongate under the swelling action of the solvent, they generate an elastic retractive
force in opposition to this deformation. The volumetric swelling reaches steady state
when the two forces balance each other. Given that the steady state swelling ratio is a
direct function of the extent of cross-linking in the sample, swelling experiments are an
effective and low-cost technique to characterize cross-linked networks. At the simplest
level of analysis, swelling measurements can be used for quality control and serve as an
indexing tool for polymer systems with different levels of cross-linking. At a higher level
of analysis, molecular weight between cross-links and the number of cross-links per
chain can be computed, if one knows the Flory interaction parameter for the polymer-
solvent system (105). With network polymers, it is common to speak of cross-link
density, which is the number of cross-linked monomer units per main chain. Cross-link

density is deﬁned mathematically as shown in equation (3.1).

_ _ln(1- v,)+ 222 + 1,0,? (3.1)
v,— V,
<I>,(v2 —v,/ 2)

 

V
In equation (3.1), I), is the cross-link density; 02 =Tf- where V0 is the volume of the

uncross-linked network and V, volume of the cross-linked network; )5, is the Flory

Interaction parameter and (I), is the molar volume of the solvent. The higher the cross-
link density, the more rigid the polymer behaves. Elastomers are characterized by having

95

a very low cross-link density; about one cross-link per 100 monomer units-together, with
highly ﬂexible main chains to allow large deformations. If the gel particles are very
small, in the range of 300-1000 microns, they are called as microgels. Microgels behave
as tightly packed spheres that can be suspended in solvents. They have attracted
considerable interest in recent years with the development of solid phase synthesis and
techniques for immobilizing catalysts.

Covalent cross-linking has certain disadvantages. Once cross-linked, a polymer cannot be
dissolved or molded. Scrap cross-linked polymer cannot be recycled. One approach to
circumvent this problem is to introduce thermally labile cross-links, that is, chemical
cross-links that break apart on heating and reform on cooling. The other approach has
been to introduce strong secondary bonding attraction between polymer chains, such that
the polymer exhibits properties of a thermosetting material while remaining
thermoplastic. In recent years, the technology of block co-polymers has been applied to
the area of physical cross-linking. The aggregations impart a signiﬁcant degree of elastic
behavior, yet the copolymers still exhibit the ﬂow properties of therrnoplastics.

In this work, biodegradable polyesters are cross-linked using free radical
initiating systems to yield gel particles, termed as “microgels”, which reinforce the main
polyester matrix improving the properties. When the polyesters are cross-linked in batch
reactors as opposed to extruders, gel particles obtained are large in size and are termed as

“macrogels”. These “macrogel” particles lead to problems during blown ﬁlm processing.

96

3.1.3 Filled Composites

Fillers are added to polymers for a variety of reasons. Reinforcing ﬁllers improve tensile
strength or tear strength. Carbon black and silica are typical examples of reinforcing
ﬁllers. Axisyrnmetric particles such as ﬁbers and ﬂakes are used to increase the modulus
and heat deﬂection temperatures and they also come under the category of reinforcing
ﬁllers. The highest reinforcement effect is, apparently, obtained from carbon ﬁbers. It is
interesting to mention that carbon ﬁbers themselves are not at all stronger than glass
ﬁbers. However, their modulus of elasticity is 3-5 times higher. The higher Young’s
modulus of composites ﬁlled with carbon ﬁbers, compared with those ﬁlled with glass
ﬁbers (at the same concentration), leads to a situation where the fraction of load
transmitted to the ﬁber is higher, whereas that received by the matrix is lower (106). The
elasticity modulus of asbestos ﬁbers is about 1.5 times higher than that of glass ﬁbers.
However, these ﬁbers have a very uneven length distribution, which reduces the
reinforcement effect. Compared to glass ﬁbers, standard organic polymer ﬁbers have a
lower Young’s modulus and provide a correspondingly lower reinforcement effect.
Natural organic ﬁbers, including cellulose, wood ﬂour etc. provides almost zero
reinforcement effect due to their small size, moisture holding capability, poor interfacial
adhesion and great ﬂexibility.

Composites may also be reinforced with lamellar ﬁllers. For these ﬁllers, the aspect
ratio is deﬁned as the ratio of linear dimension to thickness. lamellar composite ﬁllers
exhibit reinforcing effect in two directions, rather than in one as in the case of ﬁbers.

Therefore, randomly distributed lamellar ﬁllers are expected to produce better

97

reinforcement effect than ﬁbers, for the same ﬁlling ratio. Common ﬁllers such as talc
and kaolin also have a lamellar structure, but do not exhibit a reinforcing effect due to
their low aspect ratios.

The behavior of inert ﬁller is illustrated by the measurements of Han on calcium
carbonate-ﬁlled polypropylene (107). The viscosity increases considerably as the ﬁller
concentration increases. Vinogradov and co-workers (108) studied the inﬂuence of
carbon black on the rheology of polyisobutylene. It was established that the addition of
carbon black reduced die swell, improving processability. In a study of the injection
molding of highly ﬁlled phenolic compounds, Beck (109) found that the mold ﬁlled by
‘jetting’; at the mold entrance, the jet emerging from the gate traveled through the mold
until it hit the opposite wall; mold ﬁlling occurred by the buckling and spreading of the
solid like jet. Hall and co-workers (110) studied the injection molding process of a 40-
percent glass ﬁber ﬁlled polycarbonate blend. It was observed that the ultimate tensile
strength varied from 22,000 psi for a small tensile bar to 11,000 psi in the square plate
perpendicular to the ﬂow direction. Various researchers (111,112) have shown that the
volume fraction of the ﬁller as well as the ﬁller size affects the ﬂow of the ﬁlled
composite in a mold. Inert ﬁllers and pigments do not change the ﬁnal properties in a
desirable direction, but make the polymer easier to mold or extrude and lower the cost of
the ﬁnal composite. However, the most important issue to be addressed is the
compatibility between the ﬁller and the matrix, which can be improved by modifying
either the ﬁller or the matrix using chemical treatments. Clay, talc and calcium carbonate
are some examples of inert ﬁllers. Brittle ﬁllers of little strength, such as perlite, will fail

in a loaded specimen. This will initiate cracks and an early specimen failure. Organic

98

ﬁllers may be divided into two groups, elastic and pseudo-plastic ones. The ﬁrst group
has been extensively used for improving the impact strength of materials. In all cases, the
added elastomer reduces the tensile strength of the composite. The second group includes
fillers prepared from plastic wastes, nutshells, wood ﬂour etc. Typical examples of plastic
wastes are powdered thermoset plastics and powdered polymer mixtures. Their effect on
the strength of the composites depends on the composition of the plastic wastes and on
the strength of the polymer matrix. As usual, the strength of the composites is reduced
compared to the pure polymers, although this negative effect may be offset by an
improvement of other properties (113).

Petroleum-based oils, fatty acids, and esters are also used as plasticizers and
lubricants. Antioxidants act as free-radical sinks and stop chain reactions in oxidation.
Curatives are essential to form network of cross-links that guarantee elasticity rather than
ﬂow. Sulfur for unsaturated and peroxides for saturated polymers are used with
auxiliaries to control reaction rate (114).

The primary objective of this work was to create lightly cross-linked (low cross-
link density) biodegradable polyester systems containing microgels, using 2, 5-Bis (tert-
butylperoxy)-2, 5-dimethylhexane (Luperox 101), a free-radical initiator, in a twin-screw
co-rotatin g extruder. The biodegradable polyesters studied were Polycaprolactone (PCL),
aliphatic polyester, and Ecoﬂex, aliphatic-aromatic co-polyester. Another important goal
of this work was to blend Ecoﬂex and cross-linked Ecoﬂex (LEcoﬂex) with talc in a
twin-screw co-rotating extruder to form blends suitable to be blown into ﬁlms with

improved mechanical properties.

99

3.2 EXPERIMENTAL DETAILS

The process requirements needed to develop an extrusion process to continuously cross-

link PCL and Ecoﬂex are as follows:

Accurate metering of the biodegradable polyester and initiator.

Adequate length of the extrusion zone to meet residence time requirements
needed to cross-link the polyesters.

Suitable screw conﬁguration in order to meet residence time requirements, melt
pumping needs and mixing requirements.

Good melt temperature control.

Reasonable product throughputs of the polymer.

Based on these requirements, extrusion process was developed to continuously cross-link

PCL and Ecoﬂex in a CENTURY twin-screw co-rotating extruder. Further, Ecoﬂex and

cross-linked Ecoﬂex were extrusion processed with tale to form blends suitable for blown

ﬁlm applications. The details of the extrusion cross-linking and blending process will be

outlined in the pages that follow. The characterization of the cross-linked polyesters and

the blends will be discussed.

3.2.1 Materials and Equipment

A twin-screw co—rotating CENTURY extruder having a length: diameter ratio (IJD) of 40

and a screw diameter of 30 mm was used for cross-linking chemistries. The extruder was

electrically heated and cooled using circulating water. The screw elements needed for the

100

screw conﬁguration were also obtained from CENTURY. The same extruder was also
used for the blending of Ecoﬂex and LEcoﬂex with tale. The auxiliary equipment
included a water bath to cool the extrudate and a pelletizer to cut the strand into small
pellets.

PCL, marketed under the trade name TONE polymers, was obtained from Dow
Chemicals. TONE polymers are homopolymers of e—caprolactone, a seven-member ring

compound. The chemical structure of the PCL polyester is shown in Figure 3-2.

F —

RC)‘ /
O/

O

 

 

Figure 3-2: Chemical structure of PCL polyester

The TONE polymer product family consists of: 1) TONE P-767 and P-787 polymers of
Number average molecular weight (Mn) of 50,000 g/mol and 80,000 g/mol respectively,
and 2) TONE P-737 and P-757 polymers of Mn 32,000 g/mol and 43,000 g/mol
respectively. The polymer used in this study was TONE P-787. Unlike most commercial
polymers, PCL polymers, as aliphatic polyesters, offer the unique property of being truly
biodegradable without the requirement of prior photo degradation. In compost
environments, TONE polymers will undergo assimilation by microorganisms.
Microorganisms of the type found in compost are essential for degradation, since no
microbial degradation of TONE polymers, either in water or in air alone, has been

observed for periods of at least a year. Thus, most applications of TONE polymers (e. g.,

101

adhesive, polymer additive, orthopedic) will not encounter problems with microbial
degradation. TONE polymers are assimilated by microorganisms and used as an energy
source in their growth, with the result that carbon dioxide (C02), water, and biomass are
produced. Intermediate metabolism products would not be expected to be harmful to
plant or animal life. This biodegradability is useful in applications such as ﬁlms for food
or yard waste compost bags, tree planting containers, and matrix systems for controlled
or slow release of pesticides, herbicides and fertilizers. TONE P-787 is miscible with a
wide variety of polymers like Poly (vinyl chloride), Polycarbonate, cellulose propionate
etc. It offers excellent melt forrnability and rigidity in the solid state. TONE P-787 will
run on most Linear Low Density Polyethylene (LLDPE) ﬁlm lines at temperatures of
around 200-250°F. 1 mil P-787 blown ﬁlm exhibits a tensile strength of around 7900 psi
and an elongation to break % value of around 400% (115).

Aliphatic-aromatic co-polyesters represent an innovative group of biodegradable
polyesters made by BASF & Eastman Chemical under the trade names Ecoﬂex and
Eastar Bio respectively, which degrade with the aid of naturally occurring enzymes.
Ecoﬂex is aliphatic-aromatic co-polyester based on the monomers adipic acid,
terephthalic acid and 1, 4-butane diol. The generic structure of Ecoﬂex is shown in Figure

3-3.

102

ii 11’ ii i
—' 'O—(CH2)—O—C—(CH2)—C O_'(CH2 — —
4 4 4

_ x- Jv

 

 

 

 

 

Figure 3-3: Chemical structure of aliphatic-aromatic copolyester

Ecoﬂex is completely biodegradable to EN 13432 as more than 90-percent of the organic
carbon in the material gets converted to carbon dioxide during a test period not exceeding

180 days. The degradation curve below (Figure 3-4) shows that Ecoﬂex liberates more

than 90-percent carbon dioxide in after only 80 days.

100

.3
Q

 

[theon €01,111 90]
N as
6 e

 

 

0 20 40 60 80 100
Time (Days)

Figure 3-4: Theoretical percent Carbon dioxide evolution of Ecoﬂex under EN

13432 test protocol.

Ecoﬂex is very compatible with natural materials such as starch, cellulose etc. Blown

ﬁlm extrusion is a particular area where Ecoﬂex shows balance in processing properties.

103

The cross-linking reactions were carried out using a free radical initiator. The
initiator was 2, 5-Bis (tert-butylperoxy)-2, 5-dimethylhexane (Luperox 101) obtained
from Aldrich Chemicals [78-63-7]. It was used as obtained.

Talc was obtained from Luzenac, Canada. The median diameter was 2.2 microns
with a speciﬁc gravity value of 2.8. Talc was dried in an oven at 100°C overnight prior to

extrusion.

3.2.2 Extrusion Cross-linking Studies

PCL and Ecoﬂex were pre-blended with Luperox 101 in a kitchen blender for 15
minutes. Luperox 101 concentrations of 0.1-wt%, 0.5-wt% and l-wt% were used in the
cross-linking of Ecoﬂex, while 0.1-wt% Luperox 101 concentration was used in the case
of PCL cross-linking. The mixture was fed to the feed port of the extruder using a
CENTURY feeder. Samples were removed from the die and in some cases from the vent
port (cross-linking of PCL and Ecoﬂex using 0.1-wt% Luperox 101). The screw
conﬁguration used for the cross-linking consisted of conveying and kneading elements

and is shown in Figure 3-5. The extrusion processing conditions are shown in Table 3-1.

104

Table 3-1: Processing Conditions for Cross-linking PCL and Ecoﬂex

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ZONE SET ACTUAL °C
ZONE] 15 24 °C
ZONE2 85 85 °C
ZONE3 125 124 °C
ZONE4 145 145 °C
ZONES 160 160 °C
ZONE6 165 177 °C
ZONE7 165 173 °C
ZONEB 165 171 °C
ZONE9 150 170 °C
DIE 145 150 °C
MELT TEMP. 154 °C
MOTOR SPEED 74 RPM
TORQUE 55-60 %
PRESSURE 90 Psi
FEEDER SPEED, CENTURY 150 (Rate = 101bs/hr) %
FEEDER SPEED, MiN i - %
FEEDER SPEED, EXT - %

 

 

 

 

105

 

 

v 1v w t 01 1.1
\V‘Lx 11...:

60160 60160 4214242142 28126121201202)

 

          

 

 

126D
vTu v V’ I ﬁfty .{J\LIT\-"‘w’1‘-"‘w (ﬂ- KW14 (2)
. will 11‘“th \\ ‘\ 8th°8
c .
K340502m2t>1294mm min) (1212612842142le (d).Ka4015120(1)
' 15.50
'quf'.’\,,/—T\‘ “/— W\ \/ lj'T- TW\ Tilt-“V?
110 \ \ \ ‘\ \ \ Wit .0
r"\.\._/’\
2812828128 6050 6050 60160 2812828114 2012014)l
I
l 120 I

Figure 3-5: Screw Configuration used for Cross-linking studies

The extruded strand was cooled using a water-bath and pelletized in line. The pellets
were dried in an oven at 75°C overnight to remove absorbed moisture before further

processing.

3.2.3 Determination of gel fraction

Cross-linked PCL (LPCL) and cross-linked Ecoﬂex (LEcoﬂex) were analyzed for their
gel fraction by Soxhlet Extraction Studies conducted in TetraHydroFuran (THF) solvent.
Around 2-3 grams of the sample was accurately weighed into a cellulose extraction

thimble (33 mm x 94 mm) and placed into the thimble holder. The evaporated solvent

106

condensed into the thimble, containing the sample. The extraction was run for 24 hours.
After extraction, the thimble along-with the sample was removed and dried to constant
weight. The ﬁnal weight of the thimble with the sample was measured from which the gel
fraction was calculated using equation (3.2), where Wf and W.- are the ﬁnal and initial

weights of sample and thimble respectively.

w
Gelfraction = —-f— (3.2)

3.2.4 Extrusion blending of Ecoﬂex with talc

The pellets of Ecoflex were dried in an oven overnight at 75°C before further downstream
processing. Ecoﬂex was mixed with talc in a kitchen blender for 115 minutes. The ratio of
Ecoﬂex: talc was 70: 30. The mixture was fed to the feed port of a twin-screw co-rotating
CENTURY extruder (IJD of 40 and screw diameter of 30 mm) extruder. The extrusion
processing conditions are shown in Table 3-2 below. The screw conﬁguration was
exactly same as the one used for the cross-linking studies as shown in Figure 3-5. The
resulting strand was cooled using a water bath and pelletized. The pellets were stored in

an oven at 75°C to remove surface moisture. The process schematic is shown in Figure

3-6.

107

Polyester + talc

 

 

Vent

r] 7 Water
///////ff////1////p+

i Die l

Pelletizer

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 3-6: Process Schematic of extrusion processing of Ecoﬂex with talc

108

Table 3-2: Extrusion Processing Conditions for Ecoﬂex blending with talc

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ZONE SET ACTUAL °C
ZONEl 15 23 °C
ZONE2 85 85 °C
ZONE3 125 125 °C
ZONE4 145 144 °C
ZONES 160 162 °C
ZONE6 165 174 °C
ZONE7 165 169 °C
ZONE8 165 167 °C
ZONE9 150 163 °C
DIE 145 147 °C
MELT TEMP. 150 °C
MOTOR SPEED 177 RPM
TORQUE 50—55 %
PRESSURE 91 Psi
FEEDER SPEED, CENTURY 150 (~12 lbs/hr) %
FEEDER SPEED, MINI %
FEEDER SPEED, EXT %

 

 

 

 

 

109

3.2.5 Thermal Analysis

A high resolution Differential Scanning Calorimeter (DSC) 2920 from TA Instruments
was used to determine the melting temperature (Tm) of the cross-linked polyesters and
their blends with talc. Nitrogen was used as the purge gas at 50 mllmin. The sample was
heated from 0°C to 200°C at the rate of 10°C per minute, cooled and re-heated again.2“‘l
scan data was plotted.

A high resolution Thermal Gravimetric Analyzer (T GA) 2950 from TA
Instruments was used to determine the degradation temperature of the cross-linked gels
along-with their blends with talc. Nitrogen was used as the purge gas with the sample
purge being 60 mllmin and the balance purge being 40 mllmin. The sample was heated to

550°C at the rate of 20°C per minute.

3.2.6 Environmental Scanning Electron Microscopy (ESEM) Studies

Films of Ecoﬂex, LEcoflex and all the blends with talc were examined under an ESEM,
Electro Scan 2020 outﬁtted with a LaB6 ﬁlament. Sample was introduced into the
Specimen chamber and vacuum (1 Torr) was applied. All measurements were taken using
a bullet detector with a working distance of 6-9 mm and a take off angle of 30°.
Accelerating voltage of 20kV was used. Various magniﬁcations ranging from 50X to

ZOOOX were studied.

110

3.2.7 Blown Film Processing

Films of the samples were made using a Killion single-screw blown ﬁlm unit. The screw
diameter was 25.4 mm with L: D ratio of 25:1. The die inner diameter was 50.8 mm with
a die gap size of 1.5 mm. The blow up ratio of the ﬁlms was 2-2.5, comparable to

Polyethylene ﬁlms. The blown ﬁlm processing conditions are shown in Table 3-3.

Table 3-3: Blown Film Processing Conditions for talc filled polyester blends

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Die 3 Die 2 Die 1 Adaptor Clamp Ring Zone 3 Zone 2 Zone 1
Sew?) 70 350 355 360 360 395 395 300
Actual (°F) 73 350 356 359 360 370 395 300
Melt (°F) 355

 

Screw Speed (RPM) 14.6

 

FPM (ft/min) 5-6

 

Pressure (psi) 1500-3000

 

 

 

 

The schematic of a blown ﬁlm process is shown in Figure 3-7.

111

 

 

 

 

 

 

 

 

 

 

 

v

Figure 3-7: Blown Film Processing Schematic

3.2.8 Mechanical Property Determination

Tensile properties of the ﬁlms were determined using INSTRON Mechanical Testing
Equipment ﬁtted with a 100 lbs load cell. The crosshead speed was 1 inch per minute.
Rectangular ﬁlm samples, S’xl’ dimension were conditioned at 25°C and 60% Relative

Humidity for 40 hours before being tested according to ASTM D-882 testing protocols.

112

3.3 RESULTS AND DISCUSSION

The cross-linking chemistry by free radical reaction is postulated in this section. The
effect of residence times and initiator concentrations on the gel fraction of cross-linked
PCL and Ecoflex (with increased molecular weight) will be discussed in this section. The
effect of various kneading elements used in the screw conﬁguration on dispersion of the
tale particles in the Ecoﬂex polyester matrix will be schematically presented and
discussed. Thermal Analysis results elucidating the effect of cross-linking and blending
on percent crystallinity, melting and degradation proﬁles of the cross-linked polyesters
and the blends will be presented. ESEM results depicting the particle sizes of the gel
phase in the cross-linked polyester matrix and talc in polyester/talc blends will be
evaluated. Finally, ﬁlm properties (tensile, tear and puncture) of the blends will be

presented.

3.3.1 Cross-linking Studies

3.3.1.1 Cross-linking Chemistry

Figure 3-8 shows the cross-linking chemistry occurring by free radical reaction using
Luperox 101. From Figure 3-8, it can be observed that branching can occur either by
intermolecular or intramolecular reaction (116). Other side reactions such as backbiting

and hydrolysis may be occurring too. However, from the tensile properties of cross-

113

linked polyester ﬁlms, we can postulate that branching is the dominant reaction. Further,
as the polyester was dried in an oven before extrusion, hydrolysis reactions can be treated

as not dominant.

Radical Coupling

lnterrnolecular reaction
0 O

WCHZ—C—Om + RC5 —> wwEQH—o—om + ROH
1,:
HC\

«ww- H— C—Om

0
www- H—g—Om \ ﬁ
1 MCHz—‘C—Om +

Cross-linking and micro-gel formation

Intramolecular reaction

MCH2——C—-om—> mCHg—C°+ O=fm —-> Branching
H /

Cross-linking and micro-gel formation
Thermohydrolysis
0 l

o Cw+H20—> moH+Ho—c«ww

11*”

Figure 3-8: Cross-linking chemistry by free radical reaction

114

3.3.1.2 Effect of Residence Time on gel fraction of cross-linked polyesters

Gel fraction results of LPCL and LEcoﬂex samples removed from the vent port (distance

of 34 D from the feed port) and the die are shown in Table 3-4.

Table 34: Gel fraction results of LPCL and LEcoflex samples obtained from the

vent and the die.

 

Polyester Gel fraction-Sample from vent Gel fraction-Sample from die

 

PCL 0.516 0.631

 

Ecoﬂex 0.282 _ 0.401

 

 

 

 

 

It is clearly observed from data in Table 3-4 that cross-linking reaction is dependent on
the residence time in the extruder. Samples removed from the vent exhibited lower values
of gel content as compared to the ones removed from the die. Moreover, LEcoﬂex
exhibited lower values of gel fraction than LPCL. This is attributed to the presence of
bulky aromatic rings in the Ecoﬂex structure, which make them less prone to free radical

chemistries.

115

3.3.1.3 Effect of Initiator Concentration on the gel fraction of cross-linked

polyesters

The effect of increasing Luperox 101 content on the gel fraction of LEcoﬂex samples

obtained from the die is shown in Table 3-5.

Table 3-5: Gel fraction results of LEcoflex containing varying amounts of Luperox

 

 

 

 

101
Luperox 101 content Gel fraction
(W t-%)
0.1 0.401
0.5 0.482
1 0.527

 

 

 

 

It can be seen that as the initiator concentration increases, gel fraction increases.
However, the rate of increase reduces as the initiator concentration increases suggesting
that the gel fraction would level off at a particular initiator concentration. These results
are in accordance with previously published reports (100,101). Takashima and Nakayama
reported that the gel content of PE cross-linked using dicumyl peroxide at 140-180°C and
then extracted using xylene exhibited higher values of gel fraction as compared to PCL or
Ecoﬂex. This could be due to the fact that PCL and Ecoflex have a lower amount of
methylene group content in the main chain as compared to PE. This makes the number of

cross-linking sites fewer. Moreover, the polymer-solvent solubility difference could also

be involved.

116

3.3.1.4 Thermal Analysis

Figure 3-9 show the DSC results of Ecoﬂex and LEcoflex (cross-linked with 0.1-wt%

Luperox 101).

 

 

 

 

 

 

 

 

    

 

 

 

Ecoflex
"""" LEcoﬂex
0.6 '
40. 8 0C
,' 66.99 0C
0.41 \
0.2. /
Heat Flow (W/g)
0.0 ‘
93.04 °C 87.81 0C
13.11 II 6.87 J/
—0.2 ' L g g
114.58 °C 123.56 0C
-0.4 , , _ , - , _ ,
'50 0 50 100 150 200
Exo Up
Temperature (°C) Universal V3.0
TA Instruments

Figure 3-9: DSC results of Ecoﬂex and cross-linked Ecoflex (0.1-wt% Luperox 101)

As seen from Figure 3-9 shown above, crystallinity was present in both cross-linked
Ecoﬂex and regular Ecoﬂex. The area under the melting peak of the Ecoﬂex polyester
was 13.11 J/g while that of [Ecoﬂex was 6.865 J/g. It is well known that the crystallinity

of a polymer reduces with the introduction of cross-linking. However, in the case of

117

cross-linking Ecoﬂex polyester using peroxide initiators, the level of cross-linking are
low such that the polymer chains have mobility to crystallize. However, as cross-linked
domains cannot crystallize and reduce crystallinity, the area under the melting peak of
LEcoﬂex was lower than Ecoflex. Moreover, the melting temperature of LEcoflex
(114.58°C) was lower than Ecoﬂex (123.56°C). This can be explained by the fact that the
size and perfectness of LEcoflex crystals are relatively low because cross-links restrict
chain motions. Further, peak crystallization temperature of LEcoflex was 26°C lower
than that of Ecoﬂex. This could be attributed to the presence of some substances acting as
nucleating agents. Humki (117) attributed the process of so-called spontaneous
nucleation to foreign substances such as catalyst residues, oxidatively degraded polymer,
or other processing determined impurities. The nucleating agent should be wetted’or
absorbed by the polymer and considering this instance; a by-product of the cross-linking
reaction could act as impurity that can initiate crystallization. Similar results were
observed when PCL was cross-linked using Luperox 101. Figure 3-10 represents the

TGA results of Ecoﬂex and LEcoflex.

118

 

- ................. Ecoflex

 

 

 

    
   

 

 

o .................. LEcoﬂex
100 ‘
80 ‘ 449.98 0C x
43.14%
60 1
Weight (%)
4O 1
449.98 0C
20 1 26.33%
0 1 u v . 1 '7 v T 1 i v 2
0 100 200 300 400 500 600
Temperature (°C) Universal V3.0
TA Instruments

Figure 3-10: TGA Results of Ecoflex and LEcoflex (cross-linked using 0.1-wt%

Luperox 101)

It is evident from Figure 3—10 that LEcoﬂex degrades faster as compared to Ecoﬂex. This
is conﬁrmed by the fact that at 450°C, 56.86% of the Ecoﬂex polymer is degraded while
73.67% of LEcoﬂex is degraded. Similar results are observed in the case of cross-linking
Of PCL. This could be attributed to thermal decomposition in the presence of a radical, as

explained by Plage and Schulten (118).

119

3.3.1.5 Environmental Scanning Electron Microscopy (ESEM) Results

ESEM images of Ecoﬂex and LEcoﬂex are shown in Figure 3-11 and Figure 3-12

respectively.

 

25 microns

Figure 3-11: ESEM Results of Ecoﬂex

120

 

25 microns I

Figure 3-12: ESEM Results of LEcoﬂex

It can be observed from Figure 3—11 and Figure 3-12 that the size of the cross-linked gels
is of the order of 1-5 microns and hence they are termed as ‘micro gels’. The gel particles
are unifomrly distributed in the polyester matrix. Thus, in this system, the cross-linked
p01yester rnicrogels reinforce the polyester matrix and provide for increased strength. The
use of an extruder (discussed in the next section 3m with its high shearing action

results in “rnicrogel” formation as opposed to macrogels.

121

3.3.2 Polyester-talc blending studies

3.3.2.1 Effect of screw configuration on extrusion processing of Ecoﬂex with talc

The screw conﬁguration used for the extrusion processing of Ecoﬂex with talc consisted
of conveying and kneading elements as shown in Figure 3-5. The main function of the
kneading blocks is to break and disperse the talc particles in the polyester matrix. The

process of dispersion is illustrated in Figure 3-13. In the ﬁgure below, (a), (b), (c) and (d)

have the same notations as represented in Figure 3-5.

M11191“?‘ .8 ‘ "S 11:11!
(a) (b) ' d)

K34 81c) (21)

ParticleA

o o o o ,

6:9 6 o O o 0 353353. 2% ° :a.

® ©® e :0 6% e g 90":0'065 30‘:

/C ® @ @Dgoe 09 90¢. 00 0:41.

@ e o 0 30° . o 4.09
ParticleB

l 2 3 4 5 6 7

 

4‘13. I"..-

    
 

 

 

 

 

 

 

1 1

 

 

 

Figure 3-13: Schematic of mixing provided by kneading blocks used for the

extrusion processing of Ecoflex with talc

122

 

Let ‘A’ and ‘B’ represent two particles present inside the extruder. KB 40/5/42 represents
a wide kneading block effective for dispersive mixing. Hence, it can be observed that the
individual particles A and B break down into many smaller particles respectively in zone
1. Zone 2 is merely a conveying zone and the respective particles are conveyed down the
length of the extruder. Zone 3 has KB 45/5/14 elements, which are small kneading blocks
effective in distributive mixing. Hence, it can be observed that particles A and B are
distributed amongst each other in zone 3. Zone 4 has a neutral kneading block
represented as KB 90/5/28. A neutral kneading block effectively accomplishes both
dispersive and distributive mixing functions. Hence, it can be observed that particles ‘A’
and ‘B’ are even further broken down and distributed in zone 4. In zones 5, 6 and 7, the

particles are further conveyed and kneaded and ﬁnally exit from the die.

3.3.2.2 Thermal Analysis Results

Figure 3-14 and Figure 3-15 represent the DSC scans of Ecoflex/talc (70/30; w/w) and

LEcoﬂex/talc (70l30; w/w) respectively.

123

 

—— Ecoﬂex

 

 

 

 

 

 

 

"""" Ecoﬂex/talc (70/30)
0-6‘ 40.88 0c
0.4. 78.94 0C
0.24
Heat Flow (W/g)
0.0‘
92.28 0C 92.47 0C
_0.21 1 3.12 J/g ‘ 13.52 J/g
_0.4 . . . T 115.68 0C . 121.46%. 1
-50 0 50 100 ~ 150 200
Exo Up
Temperature (°C) Universal V3 .0
TA Instruments

Figure 3-14: DSC Results of Ecoflex/talc (70l30; w/w)

124

 

—— LEcoﬂex

 

 

 

 

 

 

"""" LEcoﬂex/talc (70/30)
0-6 66.99 0c
04 1 74.95 °C
0.2 ‘
Heat Flow (W/g)
0.0 ‘
H<"'""’
86.69 0C
_0.4 . r 114.58 0C . 3.73 J/g
-50 0 50 100 150 200
Exo Up
Temperature (°C) Universal V3.0
TA Instruments

Figure 3-15: DSC Results of LEcoflex/talc (70/30; w/w) [LEcoflex made with 0.1-

wt% Luperox 101]

It is observed from Figure 3-14 and Figure 3—15 that blending with talc reduces the
crystallinity of both Ecoﬂex and LEcoﬂex. Thus the area under the melting peak of
Ecoﬂex reduces from 13.52 J/g to 3.126 J/g for the Ecoﬂex/talc blend. Similarly, the
crystallinity of LEcoﬂex reduces from 7.5 J/g to 3.727 J/g for the LEcoﬂex/talc blend.
This effect is similar to the one observed during the cross-linking of Ecoﬂex. This is
explained by the fact that addition of talc particles disrupts the orderliness and hence
brings down the crystallinity. Moreover, this crystallization process is enhanced in the

presence of talc particles, which act as nucleating agents. This is the reason for the peak

125

crystallization temperature of the Ecoﬂex/talc blend being 78.94°C while that of pure
Ecoﬂex being 40.88°C. It can also be observed that the melting temperature slightly
reduces on blending with talc. Thus, while the Tm reduces by 8°C for the Ecoﬂex/talc
blend as compared to pure Ecoﬂex, it reduces by about 1°C for the LEcoflex/talc blend.
This is again due to the fact that the size and perfectness of the crystalline domains in the
blends are low. Thus, while presence of talc increases the crystallization rates, it brings
down the overall crystallinity of the Ecoﬂex or LEcoﬂex. Figure 3-16 and Figure 3-17

represent the TGA scans of Ecoﬂex/talc (70/30; w/w) and LEcoﬂex/talc (70/30; w/w)

 

 

 

 

 
    

respectively.
0 ................. Ecoflex,
n .................. Ecoﬂex/talc (70/30)
100.
80 . 449.98 0C
‘ 43.14%
Weight (%)60‘ 449.98 0c
40 1 42.88%
20 4

 

 

0 100 200 300 400 500 600
Temperature (°C) Universal V3.0
TA Instruments

Figure 3-16: TGA Results of Ecoﬂex/talc (70/30; w/w)

126

 

- ................. LEcoflex
a .................. LEcoﬂex/talc (70/301

 

 

 

80‘

   
 

449.98 °C

, 6O
we‘gh‘ (%) 43.36%

40‘ 449.98 0c
20. 26.33%

 

0 100 200 300 400 500 600

 

Temperature (°C) Universal V3.0
TA Instruments

Figure 3-17: TGA Results of LEcoflex/talc (70l30; wlw) [LEcoflex made with 0.1-

wt% Luperox 101]

From Figure 3-16, it can be observed that the degradation proﬁle of Ecoﬂex is very
similar to Ecoﬂex/talc. This is evident by the fact that 56.86% Ecoflex has degraded
while 57.12% of Ecoﬂex/talc has degraded. From Figure 3-17, it can be observed that the
degradation proﬁle of LEcoflex is also nearly similar to LEcoflex/talc. At a temperature
of 450°C, while 73.67% of LEcoﬂex was degraded, only 81% of LEcoflex/talc was
degraded. This slight difference in data can be ignored due to processing and instrument

CITOI'S.

127

3.3.2.3 Environmental Scanning Electron Microscopy (ESEM) Results

ESEM results of Ecoﬂex/talc (70/30; wlw) and LEcoﬂex/talc (70/30; w/w) are shown in

Figure 3-18 and Figure 3—19.

 

100 microns

Figure 3-18: ESEM Results of Ecoﬂex/talc (70/30; wlw)

128

 

100 microns

Figure 3-19: ESEM Results of LEcoflex/talc (70/30; wlw)

It can be observed from Figure 3-18 and Figure 3-19 that the talc particles are uniformly
distributed in the polyester matrix. The size of the talc particles is less than 10 microns,
which makes the visual appearance of ﬁlms made with these blends devoid of particles.

Moreover ﬁlm processing is very stable and no necking issues are observed.

129

3.3.2.4 Mechanical Property Results

The tensile test results of ﬁlms made from Ecoﬂex/talc blends are shown in Figure 3-20,

Figure 3-21 and Figure 3-22.

5000‘

i

I Machine Direction
D Transverse Direction

Tensile Strength (psi)

 

 

 

 

 

 

 

I I —'
Ecoﬂex LEcoﬂex Ecoﬂexltalc LEcoflex/talc
(70/30) (70/30)

Figure 3-20: Tensile strength of Ecoﬂex and LEcoflex blends with talc

From Figure 3-20, it can be clearly seen that the tensile strength values of LEcoﬂex and
Ecoﬂex are comparable. However, LEcoﬂex/talc blend exhibits improved tensile strength
as compared to Ecoflex/talc blend in the Machine Direction. This is attributed to the
presence of “microgel” reinforcing particles, having the same chemical nature as the
polyester matrix, thereby leading to a two-phase system with improved properties.

130

Further, in the Transverse direction, the values are comparable. Tensile strength of
Ecoﬂexltalc is lower than pure Ecoflex. This is due to the fact that talc acts as inert ﬁller,
lowering the properties. Finally, it must be noted that the tensile strength of LEcoﬂex/talc

is comparable to LEcoﬂex.

45000-
400004 qp

350001

 

 

25°00 1 I Machine Direction

4 _ - ElTransverse Direction

 

 

 

20000-l

 

isoooJ

Modulus or Elasticity (psi)

100004

 

 

 

 

h...
Ecoﬂex LEcoflex Ecoﬂexltalc LEcoﬂex/talc
(711/30) (711/30)

Figure 3-21: Modulus of Elasticity of Ecoﬂex and LEcoflex blends with talc

From Figure 3-21, it can be noted that modulus increases on addition of ﬁller. Thus,
modulus of both Ecoﬂex/talc and LEcoflex/talc are at least twice higher than plain
Ecoflex and LEcoﬂex respectively. Typically, modulus values are higher in the

transverse direction as observed in Ecoflex/talc blend.

131

1200 "

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1000 -
{2‘ 8001
1::
8
a T I Machine Direction
g0 600 - 1
3 El 'IYansverse Direction
tn}
1 .0. .
n:
200 r
0 ' _..
Ecoflex LEcoflex Ecoflexltalc LEcoflex/talc
(70130) (70/30)

Figure 3-22: Break Elongation of Ecoﬂex and LEcoflex blends with talc

From Figure 3-22, it can be observed that in the machine direction, the break elongation
is relatively unaffected on addition of talc. However, in the transverse direction,
Ecoﬂex/talc and LEcoﬂex/talc show lower values of break elongation as compared to
plain Ecoflex and LEcoﬂex respectively. This is due to the good dispersion and
orientation of the ﬁllers in the machine direction. Moreover, LEcoﬂex and LEcoﬂex/talc
show lower values of break elongation as compared to Ecoflex and Ecoflex/talc

respectively.

132

The tear and puncture properties of Ecoﬂex/talc (70/30) blend are explained in Table 3-6

and compared with PE values.

Table 3-6: Tear and Puncture properties of Ecoﬂex/talc (70/30) blend

 

Material

 

 

 

Thickness Tear (g) Tear (g) Puncture Puncture

(in) MD CMD Max. (lbf) Ext. (in)

ASTM D ASTM D ASTM F ASTM F

1922 1922 1306 1306
LDPE 0.0010- 100-300 - 1.5-3.0 -
0.0015
Ecoflex/talc 0.0010- 314.6 485.8 1.47 0.59
(70/30) 0.0015

 

 

 

 

 

 

 

It can be observed from Table 3-6 that tear and puncture properties of Ecoﬂex/talc blend
are comparable or better than LDPE. This makes these ﬁlms a good replacement for

LDPE in applications such as lawn and leaf compost bags, retail carryout bags etc.

133

Chapter 4

 

POLYESTER-STARCH GRAFI‘ COPOLYNIERS BY REACTIVE EXTRUSION

PROCESSING

 

4.1 BACKGROUND

This section of the work deals with the synthesis of Ecoﬂex-Starch graft copolymers by
reactive extrusion polymerization. Such compositions were deemed suitable for
fabricating biodegradable articles like moldings, sheets and ﬁlms. Both starch and
Ecoﬂex are biodegradable macromolecules and the use of starch can control
biodegradation, provide renewable resource utilization and reduce cost. The following

aspects will be discussed:

134

. Ecoﬂex-Thermoplastic starch (TPS) graft copolymer
0 Synthesis of Ecoflex-TPS graft copolymer using maleic acid as a
transesteriﬁcation catalyst in a twin-screw extruder.
0 Characterization of the graft copolymer.
0 Mechanical properties of ﬁlms.
. Ecoﬂex-Maleated Thermoplastic Starch (MT‘PS) graft copolymers.
o Maleation of starch in a twin-screw extruder using maleic acid modiﬁer in
the presence of glycerol plasticizer to form starch maleates.
0 Characterization of MTPS
0 Synthesis of Ecoﬂex-MTPS graft copolymers by extrusion processing.
0 Characterization of the reactive blends.

0 Mechanical properties of ﬁlms.

The study speciﬁcally addressed the development of Ecoﬂex-TPS graft copolymers with
a focus on the issue of generating covalent linkages between the ester functionalities in
Ecoﬂex and the hydroxyl functionalities in starch. These graft copolymers can be further
used to produce viable compositions with useful performance properties. Further, the
study focused on modiﬁcation of starch using maleic anhydride/maleic acid to generate
functional groups. Synthesis of Ecoﬂex-MTPS graft copolymers by extrusion processing

was studied. Performance properties of such blends were determined

135

4.1.1 Starch Theory

Starch is the second most abundant carbohydrate in plants after cellulose. It is the main
storage polysaccharide in plants and is similar in structure and function to glycogen,
which is the main storage polysaccharide in animals. Main sources of commercial starch
are maize, potato, wheat, cassava and waxy maize (123). Starch is an anhydroglucose
polymer and consists of two distinct molecules, amylose and amylopectin, both
containing a-D-glucose units. Amylose is a linear or sparsely branched polymer of a
molecular weight in the range of 105 to 10° g/mol linked primarily by 01—) 4 glucosidic

linkages (Figure 4-1). The chains form a spiral-shaped single or double helix (124).

3 2 .

Figure 4-1: Structure of amylose

In contrast, amylopectin is highly multiple branched with a molecular mass of 107 to 109

g/mol. Amylopectin also contains a1-—) 4 glucosidic linkages, but has in addition

136

all—)6 glucosidic branching points occurring every 25-30 glucose units. Figure 4—2

shows the structure of amylopectin.

 

Figure 4-2: Structure of amylopectin

The native starches differ in water content and the ratio of amylose to amylopectin. The
amylose contents of most starches, such as wheat, maize and potato starch, are all in the
range of 20-30%. However, in waxy starches, the amylose content is as low as 1% (125).
Starch occurs naturally in granules. The granules vary in size and shape, depending upon
species, and can range from approximately 1 micron to as large as 120 microns. The
starch molecules are oriented radially in the granule to form roughly spherical shells. The
radial nature of the organization gives rise to the granules’ birefringence, resulting in the
characteristic Maltese cross when observed under cross-polarized light. Wide -angle and

small—angle X-ray scattering (WAXS and SAXS) have shown that starch is partially

137

crystalline. The overall crystallinity of the native starches is about 20—45% (126). The
type of native, crystalline structure, labeled as A-, B- or C-type, depends on the starch
source. Native A- and B-type crystal lattices consist of double helical, six-fold structures.
The difference between the two types is the packing density of the double helices in the
unit cell. The B-type structure is described as a more loosely packed hexagonal assembly
of the helices with a column of water molecules present in the center of the hexagonal
arrangement, whereas in the A-type structure, this column of water is replaced by a
double helix. The C-type structure is thought of to be an intermediate between the A and
B type structures (127).

As starch-water slurry is heated, a characteristic temperature is reached at which
point the granules start to swell rapidly and take in water. This is known as the
gelatinization temperature T (G). Below T (G), optical properties such as the
birefringence and X-ray pattern of the starch granules are retained. Irreversible swelling
of the starch granules occurs when an aqueous starch suspension is heated above T (G).
Water penetrates the amorphous regions consisting of amylose and the branching points
of amylopectin. During this process, amylose, the more soluble fraction of the starch
granule, leaches into the surrounding water where it can eventually form a gel upon
cooling (128). The disruption of the crystalline regions is followed by a further uncoiling
of the double helices until the granular spherulite structure is partially or completely

destroyed. Figure 4-3 represents diagrammatically the swelling process.

138

 

Native granules Swollen granules

Figure 4-3: Swelling of Starch granules

The gelatinization and melting process are dependent on the water content. A single peak,
denoted as the gelatinization endotherm (G), is observed at water content above 66%
(wlw). At water content below 66%, a shoulder appears at higher temperatures, denoted
as a melting endotherm (M). At very low water contents, the G endotherm disappears and
only the M endotherm is observed (129,130).

As the starch slurry is continually heated, the starch paste viscosity will begin to
drop. Also, excessive shear of a starch paste will disrupt the swollen granules and
eventually cause the viscosity to drop. This type of starch termed as “destructurized
starch” (DSR) can be used for the production of several useful biomaterials. US. Patent
No 5,095,054 to Lay et al., discloses the use of water as a plasticizer for starch in order to
make the material processable in for example an extruder. The applicability of this starch
is limited because of degradation due to water loss at elevated temperatures. At higher
water content, the vapor pressure of water exceeds atmospheric pressure and water
escapes leaving a brittle foamy material with a T3 above processing and decomposition
temperatures with bad thermoplastic properties. The behavior of DSR is glassy and

materials can only be processed by the addition of water or other plasticizers.

139

Thermoplastic starch (1' PS) is also described as substantially amorphous starch.
TPS is produced from granular starch by employing heat and mechanical energy in the
presence of several plasticizers, which do not evaporate substantially during processing at
high temperatures and pressure. Some of the common plasticizers used are glycerol,
ethylene glycol, polyols etc. From the description above, it has to be concluded that the
difference between DSR and TPS is still not well deﬁned. DS is also entitled as TPS
when it can be processed by any of the known thermoplastic processes such as extrusion,
injection molding, blow molding and compression molding into shaped forms without the
addition of extra plasticizers.

Plasticization of starch in an extruder using glycerol as
the plasticizer breaks the hydrogen bonds, and disrupts the granular crystalline
organization. It further releases the amorphous polymer chains with a1—> 4 and 011—) 6
linkages. Figure 44 and Figure 4-5 shows the ESEM pictures of starch and plasticized

starch respectively.

140

 

 

 

Figure 4-5: ESEM image of thermoplastic starch

141

X-ray diffractogram of regular corn starch and plasticized starch is shown in Figure 4-6.

25001.

(a) Corn starch
2000. 1 (b) Thermoplastic starch
1500’ .14

 

Count

 

 

 

0 10 16 22 28 34 40

Angle (20)

Figure 4-6: X-Ray diffractogram results of starch and thermoplastic starch

As shown in Figure 4—6 (a), two single peaks at 15° and 23° and the double peak at 17° is
characteiistic of A type crystals in starch. On plasticization, the characteristic starch
peaks disappeared resulting in new peaks at 13° and 21°. These new peaks obtained are

evidence of crystalline domains formed by retrogradation of the amorphous chains (131).

142

4.1.2 Starch blends with polyesters

Due to the biodegradable nature of starch, many researchers have attempted to
incorporate starch into a variety of materials in order to improve the environmental
desirability and reduce the cost of such materials. Starch may be added as inert ﬁller
(132,133), typically in its native, unmodiﬁed state, which is generally water insoluble,
granular material. Ramsay et al. (134) studied blends of granular starch with P (I-IB-co-
HV). The inclusion of 25-wt% granular starch was reported to result in a composition
with a tensile strength of about 60% the original. The authors acknowledged that the use
of unmodiﬁed granular starch as a particulate ﬁller did not offer any appreciable
reinforcement due to the poor adhesion of the polymer granule interface. In such cases,
the starch granules will normally behave as any other solid particulate ﬁller and will
contribute little, if any, in terms of improving the mechanical properties of the resulting
material. The thermodynamic incompatibility between starch and synthetic polymers
leads to the poor performance properties of these blends. Alternatively, starch that has
been gelatinized dried, and then ground into a powder may also be added as particulate
ﬁller. Although starch may be added as ﬁller, its more interesting and technologically
challenging uses have been in the area of using starch as a binder, as a thermoplastically
processable constituent within thermoplastic polymer blends, and as a thermoplastic
material by itself. While native starch does not typically behave as a thermoplastic
material by itself, it is thermoplastic in the presence of a plasticizer when heated and
sheared. Examples of patents that disclose the manufacture of “destructurized starch” and

blends of destructurized starch and other polymers include US. Patent 4,673,438 to

143

Wittwer et al.; U.S. Patent 4,095,054 to Lay et al.; US. Patent 5,256,711 to Tokiwa et al.;
US. Patent 5,275,774 to Bahr et al.; US. Patent 5,382,611 to Stepto et al. and US. Patent
5,405,564 to Stepto et al. Lately, there have been considerable publications in the use of
thermoplastic starch (TPS) as a component in multi phase blends (135-139). Still others
have manufactured thermoplastic starch blends in which native starch is initially blended
with a small quantity of water and a less volatile plasticizer such as glycerin in order to
form starch melts that are subjected to a degassing procedure prior to cooling and
solidiﬁcation in order to remove substantially all of the water therefrom. Examples of
such patents include US. Patents 5,412,005, 5,280,055, 5,288,765, 5,262,458, 5,462,980
and 5,512,378 to Bastioli et al. Although many have attempted for years to discover the
“perfect” starch/polymer blend that would yield an environmentally sound polymer
while, at the same time, fulﬁlling desired mechanical and cost criteria, such a
combination has been difﬁcult to achieve. The reason for this is that the emphasis has
been on ﬁnding the optimal polymer or mixture of polymers and other admixtures in
order to thereby “optimize” the properties of the starch/polymer blend. One drawback is
that most of the polymers and other admixtures are themselves signiﬁcantly more
expensive than starch, which tends to increase the cost of such polymer blends compared
to starch melts. Another drawback is that such additives wi11 only be able to marginally
alter the mechanical properties of the starch/polymer blends when viewed from a
materials science perspective. Hence the thermodynamic incompatibility between the
starch and the polyester at the molecular level leads to the poor performance properties of

such materials. The theory underlining incompatible blends is discussed in more detail in

144

the next section. There is also considerable discussion related to processes and methods

needed to achieve efﬁcient compatibilization.

4.1.2.1 Incompatible Blends

Polymer blends have become an important subject of scientiﬁc investigation in recent
years because of their growing commercial importance (140). Broadly classiﬁed, the two
types of polymer blends are homogeneous (i.e., miscible or compatible) blends and
heterogeneous (i.e., immiscible or incompatible) blends. In the case of homogeneous
blends, this miscibility of the two polymers results in a single glass transition temperature
(T g), for the blend, which is composition dependent, as per the Fox-Flory theory. The Fox

relationship for a two-component blend (141) is explained in equation (4.1) as follows:

= 9: + “’2 (4.1)
TSl

_1- _
T8 T82

Here T8 and T are the glass transition temperatures of the components 1 and 2
i 82

respectively, and (I)l and (132 are the volume fractions of components 1 and 2 in the

blend respectively.

Heterogeneous polymer blends, on the other hand, have many interrelated
variables that affect their rheological behavior, processability, and the
mechanical/physical properties of the ﬁnished product. For instance, the method of blend
preparation controls the morphology of the blend (e.g., the state of dispersion, the size of
the dispersed phase, and the dispersed phase size distribution), which in turn controls the

rheological properties of the blend. On the other hand, the rheological properties strongly

145

dictate the choice of processing conditions, which in turn strongly inﬂuence the
morphology.

Most of the polymer pairs are not thermodynamically miscible and so exist in two
different phases in the polymer blend. This breakdown into two phases creates an
interface, which might lead to poor performance of the blend system. Poor adhesion also
produces very low mechanical properties due to poor stress transfer between phases and
hinders the formation of highly structured morphologies. The interaction between the
polymer pairs at the molecular level has been deﬁned by the term “compatibility”.
Compatibility is treated as a relative term and can be deﬁned by a compatibility number
as explained in equation (4.2).

NC = (Experimental Probe size )/ (Domain sizes of phases) (4 2)

The experimental probe size can be taken as the scale of resolution of an instrumental
technique. The domain size is the average dimension of the dispersed phase in the blend.

Thus, when NC —> oo , the system is compatible; when NC —-) 1, and when NC —-) 0, the

system is incompatible. (142, 143, 144).

Incompatible polymer blends can be organized into a variety of morphologies.
Many properties, and subsequently uses, of a blend depend critically on the nature of this
arrangement of the two phases. One phase may be dispersed in a matrix of the other, and
in this case the matrix phase dominates the properties. A parallel arrangement allows both
phases to contribute to many properties in direct proportion to their composition in the
blend, but this is a nonisotropic structure since perpendicular to this direction the system
may represent a series arrangement with properties that disproportionately favor one

phase. Most incompatible blends investigated to date show phase separation. The phases,

146

however, vary in amount, size, sharpness of their interfaces, and degree of continuity.
The type of morphology depends upon several factors, the most important being the
composition and the viscosities of each component. Figure 4-7 shows the effect of those
two parameters on the morphology of the blend. The component, which is in higher

proportion or is less viscous, tends to form the continuous phase (145).

Continuous Phase
of A

Viscosity ratio
A/B

Continuous Phase
of B

 

Volume Ratio

Figure 4-7: Effect of composition and viscosity on phase morphology

147

4.1.2.2 Compatibilization Theory

Simple blending of immiscible blends does not generally give a material with desired
characteristics, because of the high interfacial tension existing between the two phases.
This leads to the following conditions:

0 Poor interfacial adhesion

0 Difﬁculty of attaining the desired degree of dispersion

0 Poor mechanical properties.

In general, two component polymer mixtures can be described by the following equation

(4.3)

P=P1wl + P2W2 + APE (w) (4.3)

In equation (4.3), P is the property of the blend, P1 and P2 are the property values of the
components, w, and W; are the mass fractions of components 1 and 2 and APE (w) is an
excess term, which is dependent on the composition under consideration. If APE (w)
equals zero, there is an additive result, and the property of the blend is the weighted
arithmetic average of the constituents’ properties. For polymer alloys, however, the
compatibilization achieved results in positive values of APE (w), which means that the
modulus of the polymer combination is better than the weighted arithmetic average of the
constituents’ properties (146). There are various routes for obtaining this effect of
compatibilization.

The primary way to improve interfacial adhesion is to use a compatibilizing agent

such as a block or graft copolymer. The compatibilizer added stabilizes the morphology

148

by lowering the interfacial tension between the two phases and thus improves the
properties of the blend. These graft or block copolymers can be formed in situ during the
blend preparation by using polymers containing reactive functional groups. The blending
is performed under the conditions that promote the reaction. This method is commonly
known as ‘reactive blending’. The reactivity of the functional groups is an important
parameter in reactive blending. Most of the blends are commercially prepared in an
extruder. The functional groups should react to form the required concentration of graft
or block copolymer in the short residence times typical of extrusion processes. From this
point of view, a cyclic anhydride group may react more quickly than the carboxylic group
because of its higher reactivity. Anhydride functionality can be incorporated into a
polymer chain by copolymerization or grafting of anhydrides like maleic anhydride. If a
reaction occurs between the compatibilizer and the blend components, it will improve
adhesion tremendously. The adhesive strength resulting from chemical bonding is about
35 times greater than that resulting from Van Der Waals attraction. In order for a reaction
to occur, the components of the blend need to be functionalized, i.e. speciﬁc segments
need to be incorporated on the polymer chain that can react with the copolymer. When
the adhesion between two polymers is due to the reaction between one of the polymers
and a functional group contained in the other, then the adhesive strength is given by the

expression as shown in equation (4.4).
f = “C” (4.4)

where f is the adhesive strength, C is the concentration of the functional group, and a and
b are positive constants. There are also examples in literature where the compatibilizer

added does not undergo reaction with the blend components (147,148). In this study, two

149

methods were employed to achieve this compatibilization as explained above. The ﬁrst
method was to reactively graft a hydrophilic polymer such as starch onto Ecoflex
polyester backbone using trans-esteriﬁcation chemistry. The second method was to
maleate starch to form maleated thermoplastic starch and then use the new maleated
starch to form graft copolymers with Ecoﬂex. Prior work in the areas of reactive blending

and starch derivatization is explained in sections 4.1.3 and 4.1.4 respectively.

4.1.3 Starch-Polyester Reactive Blends

As explained in the previous sections, the thermodynamic incompatibility between the
starch and the polyester often leads to poor performance of these blends. In order to
improve this drawback, graft copolymerization of vinyl monomer on the starch backbone
was used to modify starch. Fanta and Bagley have reviewed the synthesis and discussed
some applications of starch graft copolymers (149,150). Otey et al. (151,152,153)
blended starch with poly (ethylene-co-acrylic acid) (EAA). In these papers, the authors
suggested the formation of hydrogen bonds between the carboxylic group in EAA and the
hydroxyl group in starch. Increasing the level of starch decreased the percentage
elongation of the ﬁlm and increased the diffusion rate of water. Similar complexes like
EAA can also be formed with the hydroxyl groups of the polyethylene-vinyl alcohol
(EVOH) copolymer. They report a reaction between the anhydride group in the synthetic
polymer with the -—OH groups of starch. Bloembergen et al. (144) have reported on
blends and alloys containing lignocelluloses like starch, cellulose acetate etc. US. Patent

5,314,934 to Tomka et al. provides a process to produce a polymer mixture essentially

150

consisting of thermoplastically processable starch (up to 70—wt %). During this mixing
process, an esteriﬁcation reaction takes place between the maleic anhydride groups in the
terpolymer with free hydroxyl groups in the starch. These reactive blends can be blown
into ﬁlm with properties comparable to LDPE. US. Patent 5,234,977 to Bastioli et al.
discloses a material used for the production of biodegradable articles in ﬁlm, sheet or
ﬁber form, which can be produced by extrusion from a molten mass that includes a
synthetic thermoplastic polymer and a destructured starch to which a boron containing
compound such as boric acid has been added. Narayan et al. (154) have grafted starch
onto aliphatic polyesters. They report on the fact that when starch is not plasticized, only
surface grafting is possible. However, when starch is plasticized, the hydroxyl groups are

freely available for reaction, thus enhancing the reactivity.

4.1.4 Starch Derivatives

The term “starch derivative” includes those modiﬁcations, which change the chemical
structure of some of the D-glucopyranosyl units in the molecule. Derivatization of starch
is generally conducted to modify the gelatinization and cooking characteristics of
granular starch, to decrease the retrogradation and gelling tendencies of amylase-
containing starches and to enhance hydrophilic or hydrophobic characteristics. Starch can
also be derivatized to increase its reactive capability in further downstream blending
operations with other biodegradable polyesters. A starch derivative is fully deﬁned by a
number of factors: plant source (corn, waxy maize, potato); amylose to amylopectin ratio;

Degree of Polymerization; type of derivative (ester, ether etc.); nature of the substituent

151

group (acetate etc.); Degree of Substitution (DS) or molar substitution (MS) and physical
form (granular, pregelatinized).

Most commercially produced derivatives have a DS,
generally less than 0.2. The DS is a measure of the average number of hydroxyl on each
D-glucopyranosyl unit, which is derivatized by substituent groups. DS is expressed as
moles of substituent per D-glucopyranosyl unit. Since the majority of the glucose units in
starch have three hydroxyl groups available for substitution, the maximum possible DS is
three. DS can be expressed using the equation (4.5) as shown below; where W is the
weight-% of the substituent (based on the total weight) and M is the molecular weight of
the substituent.

162*W (4 5)

0.5.:
lOO*M-(M—l)*W

 

Modiﬁed starch products have low DS levels. They were primarily designed to alter their
solution properties for food applications or adhesion to paper. Acetylated starches, for
example, have been known for more than 100 years. Starch triacetates were prepared
earlier by carrying out the acetylation in acetic-anhydride-pyridine solution. This
procedure was impractical to scale up due to the high costs involved with pretreatment
and recovery of pyridine (155). More recently (156), cornstarch was fully acetylated by
reacting with fourfold quantities of acetic anhydride at 123°C for 5 hours. The reaction
was carried out in an alkaline medium of sodium hydroxide. The solubility of these starch
acetates is dependent upon the acetyl value, the degree of degradation, polymerization of
the ester, the type of starch and the nature of the solvent. Undegraded starch acetates are
soluble in relatively few solvents, although many reagents cause the formation of gels.

Acetyl derivatives of starch give molded products, which are brittle, clear and

152

transparent. Dispersions of triacetates in dichloromethane gave clear and flexible ﬁlms
with tensile strengths in the range of 3500 psi (157).

Highly acetylated starches, historically, were of some
interest because of their organic solvent solubility and their thermoplasticity for ﬁlm and
ﬁber applications analogous to thermoplastic cellulose esters. In spite of this
development, high DS starch esters have not been developed commercially because they
could not compete with similar cellulose derivatives in terms of strength and cost. Such
high DS starch esters are characterized by their crystalline properties exhibiting clear
melt transitions (144).

Cyclic dibasic acid anhydrides such as succinic
anhydride yield starch esters containing a free carboxylate group that increases the water-
holding capacity of the product (158). The same chemistry could be carried out with
maleic anhydride to yield the maleate half ester. The reaction schemes are depicted below
in Figure 4-8. Treatment of a starch suspension with a cyclic dicarboxylic acid anhydride
containing a hydrophobic substituent group yields products with emulsion stabilizing

properties (159).

R- H—4 - 1' II - +
Starch—0H + (1’ o BLStarch—o—c—CHz—CH—c—o Na
HZC—< pH=8 FL

0

Figure 4-8: Starch Reaction with half-esters of dicarboxylic acids

153

Most of the modiﬁcations of starch using cyclic anhydrides are disclosed in patents. This
intermediate reaction product was further heated, after the addition of succinic anhydride,
for 2 hours at the same temperature to give starch propionate succinate. These starch
esters have value as dispersants in coatings replacing commercially available paint
containing dispersants. US. Patent 3,732,207 discusses the production of starch esters by
heating starch with maleic or succinic anhydride at 100-155 °C in the presence of 0.5-
15% moisture to give a product with degree of substitution (DS) 0.02 to 0.04.

While all of the above work is related to the production of starch esters in batch,
starches can be acylated on extrusion with cyclic anhydrides in the presence of carbonate
buffers to yield the corresponding starch esters. Tomasik et al. (160) reacted corn starches
in extruders containing varying amounts of moisture (18,20 and 30%) with succinic,
maleic and phthalic anhydrides. Carbonate buffer, either pH 8 or pH 9 was added as a
medium during extrusion. It is demonstrated that extrusion of starch with cyclic
anhydrides in alkaline medium presents a facile method of preparation of anionic starches
of hydrophobic character based. However, based on up-to-date collected experiences, this
method apart from some advantages seems to have several limitations. They result from
the chemistry of the process as well as from construction of extruder. The fast, cheap,
clean, waste less process together with a possibility of a large production of such kind
starches is an advantage of this method. However, due to the high viscosity of the
thermoplastic (TPS) melt, it is extremely difﬁcult to extrude samples containing less than
25% moisture at temperatures of around 130°C. In all of the above studies, grafting of

reactive moieties onto the starch backbone in the melt phase has not been conducted.

154

In this study, starch was maleated in the melt phase using maleic
anhydride/maleic acid in a twin-screw co-rotating extruder in the presence of glycerol
plasticizer. These starch maleates were further reactively blended with Ecoﬂex polyester
in the melt phase to form Ecoﬂex-Maleated Thermoplastic Starch graft copolymers to be

used in blown ﬁlm applications.

4.2 EXPERIMENTAL

Biodegradable graft copolymers of Ecoﬂex and starch were synthesized by reactive
extrusion processing. The starch component in the copolymer was Thermoplastic Starch
(TPS) in one case and Maleated Thermoplastic Starch (MTPS) in another. Both TPS and
MTPS were synthesized in a twin-screw co-rotating extruder using glycerol plasticizer.
The details of the experimental protocols to make TPS, MTPS and the graft copolymers
will be elucidated. Further, characterization of the products along with the mechanical

properties of ﬁlms made from the graft copolymers will be discussed.

4.2.1 Materials and Equipment

Granular starch was obtained from Cargill-grade SMP 1100, un-modiﬁed com-starch
with equilibrium moisture content of about 12 percent (wlw). Ecoﬂex polyester [poly
butylene (adipate-co-terephthalate)] was obtained from BASF. Anhydrous glycerol

[2136-03], 99.9% assay was obtained from J.T. Baker. Maleic anhydride or 2, 5

155

Furandione [108-31-6] was obtained from Sigma. Maleic acid was obtained from J.T.
Baker. Luperox 101 [78-63-7] was obtained from Aldrich Chemicals. All the above
materials were used as obtained.

A twin-screw co—rotating CENTURY extruder having a length: diameter ratio
(LID) of 40 and a screw diameter of 30 mm was used for the plasticization as well as the
reactive blending chemistries. The extruder was electrically heated and cooled using
circulating water. The screw elements needed for the screw conﬁguration were also
obtained from CENTURY. The auxiliary equipment included a water bath to cool the

extrudate and a pelletizer to cut the strand into small pellets.

4.2.2 Ecoﬂex-Thermoplastic Starch graft copolymers

In this section, experimental details on reactive extrusion processing of Ecoﬂex—TPS graft

copolymers along-with the characterization by Soxhlet extraction are outlined.

4.2.2.1 Synthesis of Ecoﬂex-TPS graft copolymers by reactive extrusion processing

The synthesis of Ecoﬂex-TPS graft copolymers was accomplished in a twin-screw co-
rotating CENTURY extruder using maleic acid as a trans-esteriﬁcation catalyst. TPS was
produced by plasticization of regular com-starch using glycerol (20-wt %) as a plasticizer
in the same extruder. Maleic acid was ground to a ﬁne powder using a mortar and pestle

and pre-blended with the Ecoﬂex polyester before being fed to the feed port of the

156

extruder. The concentration of maleic acid used was l-wt% with respect to the total
concentration. Meanwhile, TPS, previously oven dried overnight at 75°C, was ground to
a ﬁne powder and fed using an external feeder to the feed port of the extruder. The feeder
rates were adjusted accordingly to obtain a ratio of 70:30 (Ecoﬂex + Maleic acid): TPS.
The temperature proﬁle and the screw conﬁguration used are shown in Table 4-1 and
Figure 4-9 respectively. In Figure 4-9, the entire screw conﬁguration is divided into 3
sections; section 1 of 12.5D distance followed by section 2 of 15.5D distance and ﬁnally
section 3 with 12D distance. The vent port was kept open to remove maleic acid and
water. The extruded strand was cooled using a water bath and pelletized in line. The

pellets were dried in an oven overnight at 75°C before being blown into a ﬁlm.

[\J‘ ‘v’ ‘\:\./

 

811181.. V V1.1

28114 6050 6050 6080 421424214228128m2000m

 

   

 

 

 

(er-K340511412)
1 10111890512831
c. m
“8"?”‘2 12) mm?” . tar-K840161211 (1)
misso '
\' _ 1V“? ”ﬁn
11111. Vimv "\
2812828128 80180 80160 80180 2812828114VVVV20120t41
l 120 1

Figure 4-9: Screw Configuration used for the synthesis of Ecoflex-TPS graft

copolymers

157

Table 4-1: Extrusion Processing conditions for the synthesis of Ecoﬂex-TPS graft

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

copolymers

ZONE SET ACTUAL UNITS
ZONEl 15 70 °C
ZONE2 95 99 °C
ZONE3 125 122 °C
ZONE4 145 149 °C
ZONES 160 170 °C
ZONE6 165 192 °C
ZONE7 165 ‘ 196 °C
ZONE8 165 193 °C
ZONE9 150 185 °C
DIE 145 153 °C
MELT TEMP. 153 °C

MOTOR SPEED 254 RPM
TORQUE 67 %

PRESSURE 92 Psia

FEEDER SPEED, CENTURY (Ecoﬂex + Maleic acid) 135 (~8 lb/hr) %
FEEDER SPEED, MINT %
FEEDER SPEED, EXT (TPS) 0.80 (3.4 lb/hr) %

 

 

 

 

 

158

4.2.2.2 Soxhlet Extraction Studies

Ecoflex-TPS graft copolymer was extracted in Dichloromethane (DCM) solvent. Around
2-3 grams of the sample was accurately weighed into a cellulose extraction thimble (33
mm x 94 mm) and placed into the thimble holder. The evaporated solvent condensed into
the thimble, containing the sample. The extraction was run for 24 hours. After extraction,
the thimble along-with the sample was removed and dried to constant weight. FTIR and
thermal analysis characterizations were conducted on the ﬁlm sample obtained by solvent

casting the extracted graft copolymer solution.

4.2.3 Ecoﬂex-Maleated thermoplastic starch graft copolymers

In this section, experimental details on maleation of starch using maleic anhydride/
maleic acid in presence of glycerol plasticizer by reactive extrusion processing are
outlined. Further, characterization techniques pertaining to analysis of MTPS are
explained. Finally, experimental details pertaining to the synthesis and characterization of

Ecoflex-MTPS graft copolymers are outlined.

4.2.3.1 Maleation of starch by reactive extrusion processing

Regular silver medal pearl com-starch, obtained from Cargill Inc. was reactively

modiﬁed using maleic anhydride/maleic acid in a twin-screw co-rotating CENTURY

159

extruder with a screw diameter of 30 mm and a LID ratio of 40. Maleic anhydride/acid
(MA) was ground to a ﬁne powder using a mortar and pestle and mixed with the starch
(800 gm starch + [20 or 40 or 80 gm] maleic anhydride/acid) using a kitchen blender for
15 minutes before being fed to the feed port of the extruder. Glycerol was mixed with
2,5-Bis (tert-butylperoxy)-2,5-dimethylhexane (Luperox 101) {200 gm glycerol + 1.6 gm
Luperox 101} and pumped to the extruder using a peristaltic pump. The relative rates of
the starch/MA mixture to the glycerol/Luperox 101 mixtures were maintained at 80:20.
The calibration curve for the glycerol pump is shown in Figure 4-10. The screw
conﬁguration used for the maleation consisted of conveying and kneading elements as
shown in Figure 4-9. Water vacuum was applied at the vent port to remove the unreacted
maleic acid/water mixture. Due to the low viscosity of the extruded strand, MTPS was
collected in aluminum pans and ground to a ﬁne powder. Ground MTPS was stored in an
oven at 50°C. The extrusion processing conditions for the reactive maleation is outlined

in Table 4-2.

160

   

y = 17.2): + 0.75
R-square = 09923

 

 

Figure 4-10: Calibration curve for Glycerol pump used in the reactive maleation of

starch

161

Table 4-2: Extrusion processing conditions for the reactive maleation of starch using

maleic anhydridelmaleic acid modiﬁer in the presence of glycerol plasticizer

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ZONE SET ACTUAL UNITS
ZONEl 15 39 °C
ZONE2 95 103 °c
ZONE3 125 122 °C
ZONE4 145 146 °c
ZONES 160 157 °C
ZONE6 165 161 0c
ZONE7 165 167 °C
ZONE8 165 164 °C
ZONE9 150 156 °C

DIE 145 147 °c
MELT TEMP. 148 °c
MOTOR SPEED 125 RPM
TORQUE 35 %
PRESSURE 92 Psia
FEEDER SPEED, CENTURY 155 (Rate = 12.3 lbs/hr) %
GLYCEROL PUMP 0.1 (Rate = 3.07 lbs/hr) %
FEEDER SPEED, EXT - %

 

 

 

 

 

162

4.2.3.2 Synthesis of Ecoflex-MTPS graft copolymers by reactive extrusion

processing

The synthesis of Ecoﬂex-MTPS graft copolymers was accomplished in a twin-screw co-
rotating CENTURY extruder. MTPS was produced by reactive plasticization of regular
com-starch using MA modiﬁer and glycerol (20-wt%) plasticizer as explained above.
MTPS, previously oven dried overnight at 75°C, was ground to a ﬁne powder and fed
using an external feeder to the feed port of the extruder. The feeder rates were adjusted
accordingly to obtain a ratio of 70:30 (Ecoﬂex: MTPS). The temperature proﬁle and the
screw conﬁguration used are shown in Table 4-3 and Figure 4-9 respectively. In Figure
4-9, the entire screw conﬁguration is divided into 3 sections; section 1 of 112.5D distance
followed by section 2 of 15.5D distance and ﬁnally section 3 with 12D distance. The vent
port was kept open to remove maleic acid and water. The extruded strand was cooled
using a water bath and pelletized in line. The pellets were dried in an oven overnight at

75°C before being blown into a ﬁlm.

163

Table 4-3: Extrusion processing conditions in the synthesis of Ecoflex-MTPS graft

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

copolymers
ZONE SET ACTUAL UNITS

ZONE] 15 37 °C
ZONE2 95 96 °C
ZONE3 125 132 °C
ZONE4 150 150 °C
ZONES 160 164 °C
ZONE6 165 176 °C
ZONE7 170 173 °C
ZONE8 170 169 °C
ZONE9 170 169 °C
DIE 150 152 °C
MELT TEMP. 155 °C

MOTOR SPEED 150 RPM
TORQUE 75 %

PRESSURE 95 Psia
FEEDER SPEED, CENTURY (Ecoﬂex) 145 (Rate = 5.5 lbs/hr) %
FEEDER SPEED, MINI %
FEEDER SPEED, EXT (MTPS) 92 (Rate = 2.4 lbs/hr) %

 

 

 

164

 

4.2.3.3 Soxhlet Extraction Studies

Ecoﬂex-MTPS graft copolymer was extracted in Dichloromethane (DCM) solvent.
Around 2-3 grams of the sample was accurately weighed into a cellulose extraction
thimble (33 mm x 94 mm) and placed into the thimble holder. The evaporated solvent
condensed into the thimble, containing the sample. The extraction was run for 24 hours.
After extraction, the thimble along-with the sample was removed and dried to constant
weight. FTIR and thermal analysis characterizations were conducted on the ﬁlm sample

obtained by solvent casting the extracted graft copolymer solution.

4.2.3.4 Fourier Transformed Infrared Spectroscopy (FTIR) Studies

FT IR analysis was conducted on samples of MTPS and the graft copolymers using Perkin
Elmer. MTPS was extracted using TI-IF in a Soxhlet extraction unit, dried and ground to a
ﬁne powder. 1% MTPS was mixed with 99% KBr powder and pressed using a pellet
maker to a ﬁne transparent disk. This disk was directly put into the sample holder in
FTIR to obtain the Spectrum. A background scan was also conducted with just the KBr
pellet, which was subtracted from the total spectrum to obtain just the sample spectrum.
Films obtained by solvent casting the graft copolymer solution were directly introduced

into the sample chamber to obtain the FTIR spectrum of the graft copolymer.

165

4.2.3.5 Thermal Analysis Studies

A high resolution Differential Scanning Calorimeter (DSC) 2920 from TA Instruments
was used to determine the melting temperature (Tm) of the MTPS and the graft
copolymer samples. Nitrogen was used as the purge gas at 50 mllmin. MT PS samples
were heated to 200 0C at the rate of 10 0C per minute, cooled to -50 0C and then reheated
to 200 0C. 2"d scan data was taken for analysis. The graft copolymer samples were heated
from 0°C to 200 0C at the rate of 10 0C per minute, cooled and re-heated again.2"'d scan
data was plotted.

A high resolution Thermal Gravimetric Analyzer (TGA) 2950 from TA
Instruments was used to determine the degradation temperature of the MTPS and the
graft copolymer samples. Nitrogen was used as the purge gas with the sample purge
being 60 mllmin and the balance purge being 40 mllmin. The samples were heated to

550°C at the rate of 20°C per minute.

4.2.3.6 Intrinsic Viscosity Studies

The intrinsic viscosity of starch, TPS and the MTPS samples were determined using a
Cannon Ubbelohde viscometer (size OB) in DMSO solvent at 25°C. Details on the

process are explained clearly in Chapter 2, section 2.2.6.

166

4.2.3.7 Blown Film Studies

Films of the graft copolymers made from both TPS and MTPS were made using a Killion
single-screw blown ﬁlm unit. The screw diameter was 25.4 mm with L: D ratio of 25:1.
The die inner diameter was 50.8 mm with a die gap size of 1.5 mm. The blown ﬁlm

processing conditions are shown in Table 4-4.

Table 4-4: Blown Film Processing Conditions for Polyester-Starch graft copolymers

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Die 3 Die 2 Die 1 Adaptor Clamp Ring Zone 3 Zone 2 Zone 1
Set (HF) 70 350 355 360 360 f 395 395 300
Actual (0F) 73 351 358 357 360 380 395 300
Melt ("F) 364

 

Screw Speed (RPM) 12.4

 

FPM (ft/min) 5-6

 

Pressure (psi) 500-1000

 

 

 

 

4.2.3.8 Mechanical Property Determination

Tensile properties of the ﬁlms were determined using INSTRON Mechanical Testing

Equipment ﬁtted with a 100 lbs load cell. The crosshead speed was 1 inch per minute.

167

Rectangular ﬁlm samples, 4’xl’ dimension were conditioned at 23°C and 50% Relative

Humidity for 40 hours before being tested according to ASTM D-882 testing.

4.3 RESULTS AND DISCUSSION

The results of various aspects of Ecoflex-TPS and Ecoﬂex-MTPS graft copolymers
discussed in previous sections will be presented. Trans-esteriﬁcation reaction chemistry
leading to the synthesis of a graft copolymer will be explained. Soxhlet extraction results
of Ecoflex-TPS graft copolymers along with the FTIR analysis and TGA results
conﬁrming the existence of a graft copolymer will be discussed. Mechanical properties of
ﬁlms, produced by blown ﬁlm processing will be presented and discussed. Processing
and characterization results of MTPS will be evaluated. Finally, results pertaining to
synthesis and characterization of Ecoﬂex-MTPS graft copolymers will be discussed. Note

that Figure 4-12, Figure 4-18, Figure 4-19 and Figure 4-28 are presented in color.

168

4.3.1 Ecoflex-TPS graft copolymers

4.3.1.1 Trans-esteriﬁcation Reaction Chemistry in the Synthesis of Ecoflex-TPS

graft copolymers

Acid-catalyzed trans-esteriﬁcation reaction chemistry is shown in Figure 4-11. Acid
catalyzed esteriﬁcations cause signiﬁcant degradation of the starch. In this particular

case, it would be expected to cause hydrolysis of the polyester in addition to the starch.

 

Figure 4-11: Acid-catalyzed trans-esterification chemistry in the synthesis of

Ecoflex-TPS graft copolymers.

169

4.3.1.2 Proof of Grafting by Soxhlet Extraction

The aim of the extraction was to selectively remove the unreacted Ecoﬂex from the
Ecoﬂex-TPS graft copolymers. However, after 24 hours of extraction, the graft
copolymer was completely extracted out and the solution in the round -bottomed ﬂask
was murky. When this solution was cast, a transparent ﬁlm was obtained. This proves
that Ecoﬂex and TPS are covalently linked. It is this covalent bonding which makes the
graft copolymer extractable in Dichloromethane (solvent wherein Ecoﬂex is soluble, but
TPS is not). However, due to the insolubility of TPS in the solvent, the graft copolymer
forms a colloidal dispersion and not a clear transparent solution. The Soxhlet extraction
results of Ecoﬂex-TPS graft copolymers are shown in Table 4-5. From Table 4-5, it is
clear that when maleic acid is not used as a catalyst, no reaction occurs, which is
conﬁrmed by the fact that the amount of polyester (Ecoﬂex or LEcoﬂex) extracted is
close to the amount of polyester initially present in the blend. This proves that the TPS
component in the blend is not attached covalently and only acts as particulate ﬁller.
However, for an Ecoflex/TPS/Maleic acid system, 97% of the material taken is extracted
out, suggesting complete reaction and formation of covalent linkages between the ester

functionalities in the Ecoflex backbone and hydroxyl groups in TPS.

170

Table 4-5: Soxhlet Extraction results of Ecoﬂex-TPS graft copolymers

 

 

 

 

 

 

 

System Polyester Starch component Amount Percent
(Ecoﬂex or (Granular starch or extracted extracted
LEcoﬂex) thermoplastic (gm) (%)
initially starch) initially
present present
(gm) (gm)
Ecoﬂex/TPS (70/30; 1.2075 0.5175 1.2595 73
wlw)
Ecoﬂex/TPS/Maleic 1.3812 0.5722 1.9074 97
acid (70l29/1)

 

 

4.3.1.3 Proof of Grafting by FTIR Analysis

The extracted Ecoﬂex-TPS graft copolymer solution was cast into a ﬁlm and the FTIR

spectrum is shown in Figure 4-12.

171

 

 

 

 

 

 

 

 

 

 

 

 

_E°°"°" ,2_5
“Ecoflex-TPS (70130)
--Ecoﬂex-TPS-Macld n ,2
(Solvent)
-—TPS
k ~15
. <
l
1 11
1 J
I
AA L: «311...... o
3400 2400 cm"1 1400 400

Figure 4-12: FTIR Spectrum of Ecoﬂex-TPS graft co-polymers

FrOm Figure 4-12, it can be observed that the graft copolymer exhibits the carbonyl
stretch peak at 1720 cm], similar to pure Ecoﬂex and the extracted Ecoﬂex/T PS blend.
Further, the ester C-O stretch at 1270 cm‘1 is observed in the graft copolymer but not in
TPS. The graft copolymer also exhibits a peak between 3200-3400 cm'1 corresponding to
—OH stretch. This peak is also observed in TPS, but not in the other two. Moreover,
primary alcohols peak (1025-1060 cm") is seen in TPS and the Ecoﬂex-TPS graft
copolymer, but not in the other two. All of the above data conﬁrm the true existence of a
graft copolymer. Finally, 5 (O-H) bend of water at 1640 cm'1 is observed in TPS and not

in any of the other materials.

172

4.3.1.4 Proof of grafting by Thermal Gravimetric (TGA) Analysis

The TGA results of solvent cast ﬁlms of extracted Ecoflex-TPS graft copolymers are

shown in Figure 4-13 as shown below.

 

' ............. Ecoﬂex/TPS
D ............. Ecoﬂex/TPS (graft co-polymer)

 

2.5 ..

 

 

2.0 1

1.5 1
Derivative weight 1

(%IOC) 1.0 -

 

0.5 .

 

0 .

 

-0.5

 

0 'ToonboV 300'400' 5'00' 600
Temperature (°C) Universal V3.03

TA Instruments

Figure 4-13: TGA Results of Ecoﬂex-TPS graft copolymers

From Figure 4-13, it is worthwhile to note that without the addition of maleic acid as a
trans-esteriﬁcation catalyst, i.e. for the Ecoﬂeme blend, no degradation corresponding

to starch is observed from the extracted portion, conﬁrming the fact that only the

173

unreactedpolyester is extracted out. However, for the Ecoflex-TPS graft copolymer,
starch degradation along-with the polyester (Ecoﬂex) is evident. This conﬁrms the
existence of a graft copolymer wherein the Ecoﬂex polyester is covalently bonded to

TPS.

4.3.1.5 Mechanical Properties Results

The tensile property results of Ecoﬂex-TPS graft copolymers are schematically shown in

Figure 4-14, Figure 4-15 and Figure 4-16.

é

ID Machine direction 1

I Transverse direction

Tensile strength (psi)

 

 

 

 

 

LDPE

 

Ecoﬂex+TPS+Maleic
acid (70/29l1)

Figure 4-14: Tensile strength results of Ecoflex-TPS graft copolymers

174

 

 

35000 " ’
CI Machine direction _.

9 Transverse direction

3 1%

15000 -

"'8
e§s

 

 

Modulus of elasticit

 

 

 

   

Ecoflex
LDPE

Ecoflex+TPS+Maleic
acrd (70/29/1)

Figure 4-15: Modulus of elasticity results of Ecoflex-TPS graft copolymers

175

 

 

 

 

 

 

 

 

600 - UMachine direction

g 500 .. HTransverse direction
.5 400 1

‘5

2‘0 300 '1

2

f; 200 -

é

m 100 '1

0 I r .

 

 

 

 

LDPE

 

Ecoflex+TPS+Maleic
acrd (70/29/1) " ' V

Figure 4-16: Break elongation results of Ecoﬂex-TPS graft copolymers

It is observed from Figure 4-14 that the tensile strength values of Ecoflex-TPS graft
copolymer, containing around 30% TPS, exhibit almost a six fold decrease as compared
to pure Ecoﬂex polyester values. This reduction in tensile strength could be attributed to
two reasons:
0 Addition of TPS, which possesses poor properties
0 Reduction in the molecular weight of Ecoflex due to acid hydrolysis, catalyzed by
maleic acid.
The tensile strength values of the graft copolymer are lower than even LDPE values as

seen from Figure 4-14.

176

 

Modulus of elasticity results, as seen from Figure 4-15, also follow the same trend as the
tensile results. It is observed from Figure 4-15 that the modulus values of the graft
copolymer are lower than both LDPE and pure Ecoﬂex. Break elongation values of the
graft copolymer (Figure 4-16) are higher than Ecoflex and LDPE. This is due to the

chemical covalent linkages, which enable stretching.

4.3.2 Maleated Thermoplastic Starch (MTPS)

This section presents the chemistry and results on reactive extrusion processing of MTPS
followed by the subsequent characterization. Characterization techniques include FTIR,

Thermal Analysis and Intrinsic viscosity measurements.

4.3.2.1 Starch Maleation chemistry

The maleation of starch using maleic anhydride is Shown in Figure 4-17. Alternatively,
maleic anhydride could react with water to form maleic acid, which could further react
with starch to form the starch maleate. During the process of maleation, there is a
reduction in the molecular weight of starch due to the hydrolysis of the starch backbone

in the presence of maleic acid.

177

 

Figure 4-17: Maleation of starch using maleic anhydride

4.3.2.2 Results on reactive extrusion processing of MTPS

The reactive modiﬁcation of regular com- starch using maleic anhydride and maleic acid
was accomplished in a twin-screw co rotating CENTURY extruder using glycerol as the
plasticizer (18-wt%) and optional Luperox 101 as a free radical initiator. When compared
to regular TPS production, the value of torque in the synthesis of MTPS was very low
(reduction in molecular weight) due to the acid hydrolysis of starch in the presence of

maleic acid. This reduction in molecular weight leads to decrease in melt viscosity values

178

 

of MT PS. Due to this, reactive maleation processing can be done at much higher
throughputs as compared to the regular thermoplastic processing of starch. Moreover,
difﬁculties in regular com- starch plasticization, such as, foaming at high throughputs or
clogging near the die are prevented during the synthesis of MTPS. However, MTPS is
difﬁcult to pelletize due to its reduced viscosity and hence use of rollers to convey the
strand is recommended. This strand could then be taken to a grinder to form MTPS

powder.

 

4.3.2.3 Proof of reaction of maleic anhydride and maleic acid onto starch by FTIR

Analysis

MTPS samples made using maleic anhydride and maleic acid were extracted in THF
solvent. Samples from the thimble and solvent (after evaporation of THF) were analyzed.

FTIR results of MTPS samples are shown in Figure 4-18 and Figure 4-19.

179

 

—Starch

 

 

 

 

P 2.5
-—TPS
—MTPS (Maleic acid)- ' 2
Solvent
—MTPS (Maleic acid)-
Thlrrble
- 1.5
<
r 1
- 0.5
I I I U I o
3900 3400 2900 2400 1900 1400

Figure 4-18: FTIR results of MTPS samples using maleic acid modifier

 

 

 

 

 

 

I 2 —- gmrch
—TPs
. 1.5 —MTPS (Malelc
anhydride) Solvent
-- MTPS (Maleic
anhydride) Thlnble
- 1
<
1 0.5
- 0
I I I I r £I5
4000 3500 3000 2500 2000 1500

Figure 4-19: FTIR results of MTPS samples using maleic anhydride modiﬁer

180

From Figure 4—18 and Figure 4-19, it can be observed that samples from the thimble and
the solvent exhibit the carbonyl stretch peak at 17 20 cm'l. This was not observed in either
the pure granular starch or TPS. This proves that the maleic acid and maleic anhydride
are covalently linked to the starch backbone. Moreover, from Figure 4—19, since the
MTPS samples show no peaks at 1780 cm'1 corresponding to ring anhydride, it is
suggested that the maleic anhydride ring is open. This is expected due to the presence of
moisture during extrusion. Further, -OH stretch peak between 3200-3400 cm'1 is

observed in all the four samples.

4.3.2.4 Thermal Analysis Results

DSC results of MTPS are shown in Figure 4-20.

181

 

 

- ............. MTPS (2.5% Maleic acid)

 

 

    

 

 

1:1 ........... Starch
+ .............. TPS
0.4 ‘
0.2 ‘
0.0 ‘ 117.33°C
Heat Flow (W/ g)
-0.2 ‘
'0'4 151.1(5)?le
140.48°C 152 870C 8
’0'6 0 50 100 ‘ 150 200 750
Temperature (°C) Universal V3.03
TA Instruments

Figure 4-20: DSC Results of MTPS samples

From Figure 4-20, it can be observed that no melting for starch is observed. This is

evident as starch degrades before it melts. However, plasticizers in the presence of heat

and shear break the hydrogen bonds in starch and destroy the crystalline nature. Thus, a

broad melting endotherm is observed in TPS, while two sharp melting endotherms are

observed in MTPS. This is attributed to the maleation of starch leading to segments

having slightly different melting characteristics. The melting temperatures of MTPS are

around 153°C MTPS and are around 10°C higher than TPS. TGA results are shown in

Figure 4—21 and Figure 4-22.

182

 

° ............. MTPS (2.5% Maleic acid)

 

 

 

El ........... Starch
120 . + .............. TPS
100 350.380C
K‘ - - ~ - . . 41.54%

801 ‘\

Weight (%) 60 1 ‘1

 

 

1 " ‘ 350.380C
40 350.380C 0. ”-09%
20 1 21.03% ,
0 100 200 300 400 500
Temperature (°C) Universal V3.03
TA Instruments

Figure 4-21: TGA Results of MTPS samples

183

 

 

 

 

 

. ............. MTPS (10% Maleic acid) , 2'0

1:1 ............. MTPS (5% Maleic acid) ' ‘ 1 1.5

+. . .. . .........MTPS (2.5% Maleic acid)
1 1.0
‘ Derivative weight (% / 0C)
. 0.5
1
t 0

5 . . . - . -2- 0.5

 

 

91

100 200 300 400 500
Temperature (°C) Universal V3.03
TA Instruments

Figure 4-22: TGA Results of MTPS samples modiﬁed with 2.5%, 5% and 10%

maleic acid

From Figure 4-21, it can be observed that MTPS is more stable than either starch or TPS.
At a temperature of 350.380C, while 75% and 63% of starch and TPS have degraded
respectively, only 58% of MTPS has degraded. Further, TPS and MTPS show additional
degradation due to glycerol.

From Figure 4-22, it can be observed that all the MTPS samples show the
derivative peak for the degradation of glycerol. Further, MTPS samples modiﬁed with 5
and 10% maleic acid Show two peaks in the derivative curve for the degradation of starch

as opposed to one peak shown by MTPS modiﬁed with 2.5% maleic acid. This is due to

184

the fact that increased maleation leads to the development of fractions having varied

degradation characteristics.

4.3.2.5 Intrinsic viscosity Results

The intrinsic viscosity of starch, TPS, MT PS (2.5% maleic anhydride and 0.1% Luperox
101) and MTPS (no LuperOx 101) were determined in DMSO solvent and are shown in
Figure 4-23 through Figure 4-26. The intrinsic viscosity of starch (Figure 4-23) was
around 1.5 dllg while that of TPS (Figure 4-24) reduced to 1.2 dllg. However, on
maleation without addition of Luperox 101, the intrinsic viscosity of MTPS (Figure 4-25)
drastically dropped to 0.27 dllg. This reduction is attributed to a decrease in molecular
weight due to the hydrolysis of starch in the presence of maleic acid. However, on
addition of Luperox 101, the intrinsic viscosity (Figure 4-26) increased back to 0.32 dllg.
This increase can be attributed to cross-linking reactions taking place, increasing the
molecular weight slightly. Further, k’-k” for the various plots were found to be close to

0.5.

185

 

 

 

c“(n/no — 1)

C"|n(n/no)

    

 

 

21 c"(n/no—-1)vs.c
1.8! y=0.97x+1.5
1.6 I
1.4 I 1

c' ln(n/no) vs. 0

1-2 ' y = -0.25x + 1.51

1 I
0.8 I
0.6 I
0.4 I
0.2 I

0 . . . .

0 0.1 0.2 0.3 0.4

Concentration ( g/dl)

 

Figure 4-23: Intrinsic viscosity of regular corn starch

 

6401/00 — 1)

0"|n(n/no)

 

c"(n/no - 1) vs. c
1.22-
1.20
1.18!
1.16:
1.140
1.12I
1.1 I
1.08I
1.06: c"ln(nln°) vs. c
1,04: y=-0.53x+1.18
1.02I

y=0.04x+ 1.19

 

 

0 0.1 0.2 0.3 0.4
Concentration ( g/dl)

 

Figure 4-24: Intrinsic viscosity of TPS

186

 

 

 

   
  

= 0.04 0.27
g 0.284 V x +
:3 0"(n/no - 1) vs. 0
'0 0.282
0.28
0.278
E 0.276 c"|n(n/n°) vs. 0
77:5 027,, I y = ooozx + 0.27
I A

 

 

0 0.1 0.2 0.3 0.4
Concentration ( g/dl)

 

 

 

Figure 4-25: Intrinsic viscosity of MTPS (2.5% Maleic anhydride)

 

c"(n/no — 1) vs. c
0.32 y = -ZE-14X + 0.32
*

6401/00 ‘ 1)
o
8

2 0.316 c"1n(nm,) vs. 0

S’
5 0315 y = -0.05x + 0.32

 

0 0.05 0.1 0.15 0.2
Concentration(gldl)

 

 

 

Figure 4-26: Intrinsic viscosity of MTPS (2.5% Maleic anhydride and 0.1 %

Luperox 101)

187

4.3.3 Ecoflex-MTPS graft copolymers

Trans-esteriﬁcation reaction chemistry leading to the synthesis of Ecoﬂex-MTPS graft
copolymers will be discussed. The extraction results of Ecoflex-MT PS graft copolymers
will be presented. FTIR analysis of ﬁlms, produced by solvent casting the extracted graft
copolymer solution will be evaluated to conﬁrm reactivity. Other characterization
techniques such as Differential Scanning Calorimetry (DSC) and Thermal Gravimetric
Analysis (TGA) will be presented. Synthesis of blown ﬁlms of Ecoﬂex-MTPS graft
copolymers, and the effect of varying blown ﬁlm pressures for different materials will be
discussed. Further, mechanical properties of ﬁlms made with the graft copolymers will be

discussed in this section.

4.3.3.1 Trans-esteriﬁcation Reaction Chemistry in the synthesis of Ecoflex-MTPS

graft copolymers

Chemistry in the synthesis of Ecoﬂex-MTPS graft co-polymers is exactly similar to the
one showed in Figure 4-11. However, a further trans-esteriﬁcation reaction between
MTPS and the hydroxy terminated polymer formed is possible and is shown in Figure

4-27.

188

ii if
CHzo—C—(l3=(|3-—C-—OH
O H H
+ MOH
—"O O_ ’-
O 0
II II
OHzo—O—<|3=cl:—c—o~wv
O H H
0” + 11,0
——0 o—
OH

Figure 4-27: Trans-esteriﬁcation reaction chemistry in the synthesis of Ecoflex-

MTPS graft copolymers.

4.3.3.2 Proof of grafting by Soxhlet Extraction

As explained in section 13_1;, the aim of the extraction was to selectively remove the
unreacted Ecoﬂex from the Ecoﬂex-MTPS graft copolymers. The Soxhlet extraction
results of Ecoﬂex-MTPS graft copolymers are shown in Table 4-6. As seen from Table
4-6, almost complete extraction is achieved for Ecoﬂex/MTPS (70/30 and 60/40) graft
copolymers. When this solution was cast, a transparent ﬁlm was obtained. This proves

that Ecoﬂex and MTPS are covalently linked making the graft copolymer extractable in

189

Dichloromethane (solvent wherein Ecoﬂex is soluble, but TPS is not). However, due to
the insolubility of TPS in the solvent, the graft copolymer forms a colloidal dispersion
and not a clear transparent solution. This result is applicable even to blends containing
MTPS made with no peroxide (see row 6 in Table 4-6). However, for the 50/50 and 40/60
blends, only 47% and 38% respectively (close to the respective Ecoﬂex amount) were
extracted out. This conﬁrms the fact that Ecoﬂex has not reacted with MTPS. Thus, the
reaction is also dependent on the relative amounts of the polyester and MTPS phases
present. Also, from Table 4-6 (rows 7 and 9), it is clearly seen that when TPS is used
instead of MTPS, there is very minimal or no reaction taking place. This suggests that the
reaction between the hydroxyl groups in starch and the ester functionalities in Ecoﬂex
occurs only in the presence of a trans-esteriﬁcation catalyst such as maleic acid. When
Ecoﬂex polyester was cross-linked and reactively blended with MTPS, around 83% of
the polyester was extracted out. This could be due to the fact that certain gel (network)

portions of the cross-linked polyester are impermeable to chemical reaction.

190

Table 4-6: Soxhlet Extraction results of Ecoflex-MTPS graft copolymers

 

 

 

 

 

 

 

 

 

 

System Polyester Starch Material %
(Ecoflex or component extracted Extracted
LEcoﬂex) (TPS or MTPS) (gm)
initially initially present
present (gm) (gm)
Ecoflex/MTPS 1.3510 0.579 1.8878 98
(70/30;w/w)
Ecoﬂex/MTPS 1.0682 0.7122 1.7010 96
(60/40;w/w)
Ecoflex/MTPS 0.7998 0.7998 0.7503 47
(50/50;wlw)
Ecoﬂex/MTPS 0.8639 0.5759 0.5513 38.2
(40/60;w/w)
Ecoflex/MTPS 1.9018 0.8150 2.6530 98
(70/30;w/w) No
Luperox 101
Ecoflex/TPS 1.2075 0.5175 1.2595 71
(70/30;w/w)
LEcoﬂex/MTPS 3.25 1.748 4.1272 83
(65/35;wlw)
LEcoﬂex/FPS 1.0225 0.4382 1.0602 73
(70/30;w/w)

 

 

 

 

 

191

 

4.3.3.3 Proof of grafting by FTIR Analysis

The extracted Ecoﬂex-MTPS graft copolymer solution was cast into a ﬁlm and the FTIR

spectrum is shown in Figure 4-28.

 

, 1.2 1 r—Ecoﬂex

-—Corn starch
1- 1
-—EcoﬂeleTPS (Maleic
anhydride)
1 0-8 1—EcoﬂexIMTPS(Malelc

 

 

acid)

 

 

 

Figure 4-28: FI‘IR Results of Ecoﬂex-MTPS graft copolymers

From Figure 4-28, it can be observed that the Ecoﬂex-MTPS graft copolymer exhibits the
carbonyl stretch peak at 1720 cm", similar to pure Ecoﬂex. Further, the ester C-O stretch
at 1270 cm'1 is observed in the graft copolymer but not in granular corn starch. The graft

copolymer also exhibits a peak between 3200—3400 cm'l corresponding to —OH stretch.

192

This peak is also observed in corn starch, but not in the other two. Moreover, primary
alcohols peak (1025-1060 cm") is seen in corn starch but not in the other three materials.
All of the above data conﬁrm the true existence of a graft copolymer. Finally, 8 (O—H)

bend of water at 1640 cm‘1 is observed in starch and not in any of the other materials.

4.3.3.4 Proof of grafting by Thermal Analysis

DSC and TGA experiments were carried out on solvent cast ﬁlm samples of the extracted

Ecoﬂex-MTPS graft copolymer. The results are shown in Figure 4-29 and Figure 4-30.

 

 

 

 

- ............. Ecoﬂex/MTPS (55/45)
a ....................... Ecoﬂex
0.6 .. 40.88°C
78.69°C

6.4. ‘
1 W ' \Ai - ‘
0.2 - V

 

 

 

 

HeatFlow (W/g) .
0.0 -
100.67°C
‘ 125.51°C
_0.2 1 _ . g _ - 5.89J/g _.
‘ 8‘.3'"” ' ' '
_0.4 - _ ﬁ -15'67 J/g ' 122.92“C
-50 0 50 100 150 200
ExoUp

Temperature (°C) Universal V3.03
TA Instruments

Figure 4-29: DSC Results of Ecoflex-MTPS graft copolymer

193

 

- ............. Ecoﬂex

 

 

 

 

    
 

 

 

E1 ............. Starch
+ .............. Ecoﬂex/MTPS (Maleic acid)
120'
330.95°C
100.
80
449.03°C
Wei t 7 60 . «
gh (a) 330.95%: ”-55%
40 ' 62.88% 449.03%:
20 . 7.05%
0 - - - - -
0 100 200 300 400 500 600

Temperature (°C) Universal V3 .03
TA Instruments

Figure 4-30: TGA Results of Ecoflex-TPS graft copolymer

From Figure 4—29, it can be observed that the melting temperature of the Ecoﬂex-MTPS
graft copolymer is very similar to pure Ecoﬂex. However, crystallinity of the graft
copolymer reduces as evinced by the area under the melting peak. Thus, the area under
the melting peak for pure Ecoﬂex is 15.67 J/g while that of Ecoﬂex-MTPS graft
copolymer is 5.89 ﬁg. This is expected as the incorporation of starch into the Ecoﬂex
matrix breaks the orderliness of Ecoﬂex bringing down the crystallinity. Moreover, peak
crystallization temperature of the graft copolymer (78.690C) is higher than that of pure
Ecoﬂex (40.880C). This is due to the presence of impurities such as catalyst residues,

which act as nucleating agents that fasten the crystallization process.

194

From Figure 4-30, it is seen that Ecoﬂex-MTPS graft copolymer exhibits two
degradations; one corresponding to the starch and the other corresponding to Ecoﬂex.
This further conﬁrms that MTPS is chemically grafted onto Ecoﬂex. Further, at a
temperature of 330°C, while 78% of the graft copolymer is un-degraded, only 63% of
starch is left behind. This is evident as MTPS is more stable than pure granular starch.
However at 450°C, while 45% of Ecoﬂex is left behind, only 7% of the graft copolymer

is left behind.

4.3.3.5 Blown Film Processing Results of Ecoﬂex-MTPS graft copolymers

Blown ﬁlms of the graft copolymers were made using a Killion single screw blown ﬁlm
unit. Temperature proﬁle in the blown ﬁlm unit, melt pressure, draw rates and blow up
ratio are some of the important processing parameters that need to be considered. Melt
temperatures greater than 420°C made the bubble raise process difﬁcult. Hence melt
temperatures below 370°C were used. At very low (less than 5 ft/min) and very high
(greater than 6.5 ft/min) draw rates; the ﬁlms were either too thick or too thin
respectively. This made the bubble instable, causing it to collapse. Hence, draw rates
around 5-6 ft/min were used. Blow up ratio was calculated to be around 2.5-3, similar to
LDPE ﬁlm processing. Blow up ratios higher than 3 led to necking issues during blown
ﬁlm processing, causing the bubble to break and collapse. The melt pressures observed

for the various graft copolymers are shown in Figure 4-31.

195

Melt Pressure (psi)

 

Figure 4-31: Blown ﬁlm pressures of Ecoflex-MTPS graft copolymers

From Figure 4-31, it is observed that the melt pressure decreases on increasing the MTPS
weight fraction in copolymer. This is attributed to a reduction in the molecular weight
(maleic acid catalyzed acid hydrolysis) of Ecoﬂex on reaction with MTPS. However, for
the same weight fraction of MTPS, melt pressure increases on cross-linking. This was
expected as cross-linking increases the molecular weight of the polyester. The melt
pressure was least for the Ecoﬂex-TPS graft copolymer and was lower than Ecoﬂex-
MTPS graft copolymer for the same TPS and MTPS weight fractions. This is attributed
to the increased reactivity of MTPS with Ecoﬂex as compared to TPS, leading to stronger

covalent linkages and increases melt pressures.

196

4.3.3.6 Mechanical Property Results of Ecoflex-MTPS graft copolymers

The mechanical properties of Ecoﬂex-MTPS graft copolymers are Shown in Figure 4-32,

Figure 4-33 and Figure 4-34.

.“

El Machine direction
I Tramverse direction

Tensile strength (psi)
.N l" 9 5" 5"
§ § § § § §

“é

 

 

Figure 4-32: Tensile Strength Results of Ecoflex-MTPS graft copolymers

197

35,000

30,000 DMachine direction
I Transverse direction
25,000

20,000 {-

15,000

 

 

10,000

  

Modulus of elasticity (psi)

5,000

 

 

 

 

 

   

Ecoﬂex
Ecoflex/MTPS
(65/35)
LEcoflex/MTPS
(65135)
LDPE

a
E
E

Ecoflex/TPS/Maleic m

Figure 4-33: Modulus of Elasticity Results of Ecoﬂex-MTPS graft copolymers

198

 

700 - L'lMachine direction
600 . .Transverse direction

 

500 d
400 "
300 '
200 '
100 '

0 .

Break elongation (%)

 

 

 

 

 

 

 

 

 

Figure 4-34: Break Elongation Results of Ecoflex-MTPS graft copolymers

It is observed from Figure 4-32 that the tensile strength values of the graft copolymer
reduce considerably as compared to Ecoflex. However, the values are only slightly less
than LDPE values. Moreover, on cross-linking, due to the increased molecular weight,
tensile strength values increase and are higher than LDPE values. The tensile strength of
Ecoflex-TPS graft copolymer is the least. This is attributed to a reduction in the
molecular weight by hydrolysis catalyzed by maleic acid.

The modulus of elasticity values (Figure 4-33) of the graft copolymers also

follows the same trend as the tensile strength values. Thus, the modulus of elasticity of

199

Ecoflex-MTPS graft copolymer is lower than pure Ecoﬂex. Further, modulus of elasticity
increases on cross-linking.

Break elongation values (Figure 4-34) of Ecoflex-MTPS graft copolymer are
higher than pure Ecoﬂex. This is attributed to chemical covalent linkages present in the
graft copolymer, making it stretchable. However, LEcoflex-MTPS graft copolymers
exhibit lower values of break elongation. This is due to a more rigid structure obtained on
cross-linking. Finally, Ecoﬂex-TPS graft copolymers exhibit the highest break elongation
values. All the values are comparable to LDPE.

The mechanical property results of Ecoflex-MTPS graft copolymers, containing
varied weight percentages of maleic acid in the MTPS, are shown in Figure 4-35, Figure

4-36 and Figure 4-37.

200

 

 

 

7,000 -

6’0“) ° C] Machine direction
A I Transverse direction
2 5,000 « *
.:
b
3 3,000 a
1:
8
t- 2’ ‘

mm- II II

.-.. I

 

 

 

 

g as; Egg Esq E
e 9'46 E‘m'a 9°? 9
0 5300 %va €300 1.:
m as “2 A3
882 83% 83%
03% 02: 0:2
as av so

Figure 4-35: Tensile Strength Results of Ecoﬂex-MTPS (70/30) graft copolymers

2.5-wt%, 5-wt% and 10-wt% maleic acid in MTPS

201

 

 

 

 

 

    

 

 

 

 

 

was
n .m E: 3282
m m so: A833
m m E92358;
.m m
m w E8 6.632
m m as 835
m m " Esacuccm
a I
rll Sea 36...:
“ smd 83c
8923:63—
lT “Seem
! u d d d c J _
m m m m m m 1
ﬁ 3 2 “D 1 5

as: 388% E 8.3.62

Figure 4-36: Modulus of Elasticity Results of Ecoﬂex-MTPS (70/30) graft

copolymers 2.5-wt%, 5-wt% and 10-wt% maleic acid in MTPS

202

 

700
1 DMachine direction
I Transverse direction

600 -

500 I

 

 

400 I

 

300

Break elongation (%)

200

100

 

 

 

 

 

 

 

 

 

LDPE

Maleic acid)
Maleic acid)
Maleic acid)

(70/30) (5%

Ecoﬂex
Ecoflex/MTPS

(70I30) (2.5%

Ecoflex/MTPS

Ecoﬂex/MTPS

(70130) (10%

Figure 4-37 : Break Elongation Results of Ecoflex-MTPS (70/30) graft copolymers

2.5-wt%, 5-wt% and 10-wt% maleic acid in MTPS

It is observed from Figure 4-35 that increase in the maleic acid content in MTPS leads to
an increase in the tensile strength value. This is attributed to increased chemical reactivity
between Ecoﬂex and MTPS. Increase in the maleic acid content leads to a greater
probability of hydrolysis reactions. However, in this case, most of the unreacted maleic
acid is removed in the vent port of the extruder, during the reactive maleation of starch,

and does not get transferred to the next stage of processing. Thus, increase in the tensile

203

strength is purely attributed to increased maleation and hence increased chemical
reactivity. However, all the values are lower than LDPE.

The modulus of elasticity values (Figure 4-36) follows the same trend as the
tensile strength values. They increase with increasing maleic acid content. All the values
observed are lower than LDPE.

Break Elongation values (Figure 4-37) of the graft copolymers are higher than
either pure Ecoﬂex or LDPE. Moreover, on increasing maleic acid content in MTPS, the
break elongation values increase. This is again attributed to the enhanced chemical
reactivity between Ecoﬂex and MTPS.

The tear and puncture properties of Ecoﬂex-MTPS (70/30) graft copolymer are explained

in Table 4—7 and compared with PE values.

Table 4-7: Tear and Puncture properties of Ecoflex-MTPS (70l30) graft copolymer

 

Material Thickness Tear (g) Tear (g) Puncture Puncture
(in) MD CMD Max. (1b,) Ext. (in)

ASTM D ASTM D ASTM F ASTM F

 

 

1922 1922 1306 1306
LDPE 0.0010- 100-300 - 1.5-3.0 -
0.0015
Ecoﬂex-MTPS 0.0010- 767.7 802.7 1 .52 0.62
(70/30) graft 0.0015

copolymer

 

 

 

 

 

 

 

 

204

It can be observed from Table 4-7 that tear and puncture properties of Ecoﬂex-MTPS
(70/30) graft copolymer are comparable or better than LDPE. Tear and puncture are
‘energy related’ properties and hence related to the crystallinity of the polymer. Hence,
an improvement in the tear and puncture properties of the graft copolymers, as compared
to LDPE, could be attributed to reduced crystallinity. This makes these ﬁlms a good
replacement for LDPE in applications such as lawn and leaf compost bags, retail carryout

bags etc.

205

Chapter 5

 

SUMMARY AND CONCLUSIONS

 

5.1 SUMNIARY

The overall scope of this work was to study and evaluate three different chemistries;
namely polymerization, cross-linking with blending, and grafting in a twin-screw co-
rotating extruder. These biobased biodegradable polymeric materials and blends
synthesized by reactive extrusion processing, have applications in the areas of
biodegradable ﬁlms, injection molded articles etc. Speciﬁcally, this work focused on the

following:

206

I Extrusion polymerization of 1, 4-dioxan-2-one (PDX) monomer and co-
polymerization with E-caprolactone (CL) monomer.

I Reactive extrusion cross-linking of biodegradable polyesters (Ecoﬂex and PCL)
and blending with talc.

I Reactive extrusion processing in the synthesis of Ecoflex-Starch graft

copolymers.

The key studies of the above areas of this work will be summarized below:

5.1.1 Extrusion polymerization of l, 4-dioxan-2-one (PDX) monomer and co-

polymerization with e-caprolactone (CL) monomer.

The extrusion polymerization of PDX monomer in the synthesis of poly (1, 4-dioxan-2-
one) (PPDX) polyester, using aluminum tri-sec butoxide (ATSB) initiator in a BAKER
PERKIN co-rotating twin-screw extruder was studied. The co-polymerization of PDX
monomer with a few mole percent CL monomer in the synthesis of poly(caprolactone-co-
1,4-dioxan-2-one) (PCL-co-PPDX), using aluminum tri-sec butoxide (ATSB) initiator in
a BAKER PERKIN co-rotating twin-screw extruder was also studied. Finally, the
extrusion polymerization process of CL monomer conducted in a BAKER PERKIN twin-
screw extruder was modeled as an Axial Dispersed Plug Flow Reactor using Residence

Time Distribution Analysis.

207

A twin-screw extrusion process to polymerize PDX monomer to high molecular

weight (up to 94,550 g/mol, each arm) PPDX polymer was designed.
The extrusion process synthesis of PPDX with theoretical number average
molecular weights ranging between 50,000 and 100,000 g/mol was targeted;

extrusion temperatures ranging from 130 to 170°C were evaluated.

A screw conﬁguration comprising entirely of conveying elements was used for

the synthesis of high molecular weight PPDX and PCL-co-PPDX polymers.

PDX monomer conversions in the synthesis of PPDX polymer were determined

by dissolution-precipitation method.

A twin-screw extrusion process to co-polymerize PDX monomer with 11, 16 and

24-mol% CL monomer at a temperature of 130°C was designed.

PDX monomer conversions in the synthesis of PCL-co-PPDX polymer were

determined by dissolution-precipitation method.

Glass transition, melting and crystallization were determined for the PPDX and

PCL-co-PPDX polymers.

208

 

o Isothermal gravimetric analysis (TGA) experiments at 180°C under Helium flow

were conducted for PPDX and PCL-co—PPDX polymers.

I Residence Time Distribution (RTD) studies were conducted during the
polymerization of CL monomer to evaluate the applicability of the Axial

Dispersed Plug Flow Reactor model to the observed RTD and conversion data.

5.1.2 Reactive extrusion cross-linking of biodegradable polyesters (Ecoflex and

PCL) and blending with talc.

Micro-gelled cross-linking of Ecoﬂex (aliphatic-aromatic co-polyester) and PCL
(aliphatic polyester) by reactive extrusion using peroxide initiator was evaluated. Blends
of cross-linked polyesters with talc were prepared by extrusion processing. These blends

were made into ﬁlms useful for lawn and leaf disposal.

I Cross-linking of Ecoflex and PCL using 2, 5-Bis (tert-butylperoxy)-2, 5-
dimethylhexane (Luperox 101) was achieved in a CENTURY twin-screw co-

rotating extruder at a temperature of around 160°C.

0 Cross-linked Ecoﬂex (LEcoﬂex) and cross-linked PCL (LPCL) were

characterized for the gel fraction by Soxhlet extraction studies.

209

5.1.3

Melting, crystallization and degradation temperatures of LEcoﬂex and LPCL

were evaluated.

LEcoflex and LPCL were characterized by Environmental Scanning Electron

Microscopy (ESEM) measurements.

Ecoﬂex and LEcoﬂex were extrusion processed with tale in a CENTURY twin-

screw extruder at a temperature of around 160°C.

Melting, crystallization and degradation temperatures of talc ﬁlled polyesters

were studied.

The talc ﬁlled polyesters were characterized by Environmental Scanning

Electron Microscopy (ESEM) measurements.

The mechanical properties of Ecoﬂex and LEcoflex blends with talc were

evaluated.

Ecoﬂex-Starch graft copolymers by reactive extrusion processing

Graft copolymers of Ecoﬂex with thermoplastic starch (TPS) and maleated thermoplastic

starch (MTPS) were evaluated. Films of these graft copolymers that were useful in

making compostable bags were studied.

210

Grafting of Ecoflex chains onto TPS by reactive extrusion processing, using

maleic acid catalyst was conducted.

Reactivity was conﬁrmed by Fourier Transformed Infra Red Spectroscopy

(FTIR) and Thermal Gravimetric Analysis (TGA).

Mechanical properties of Ecoﬂex-TPS graft copolymers were evaluated.

Maleation of starch in a twin-screw extruder using maleic anhydride or maleic

acid modiﬁers and glycerol plasticizers with optional radical initiator was

studied.

Synthesis of starch maleates was conﬁrmed by FTIR analysis.

Intrinsic viscosity experiments were carried out to document the reduction in

molecular weight and viscosity on maleation.

Melting and degradation temperatures of MTPS samples were evaluated by

Thermal Analysis experiments.

Grafting of Ecoﬂex chains onto MTPS by reactive extrusion processing was

conducted.

211

0 Reactivity was conﬁrmed by Fourier Transformed Infra Red Spectroscopy

(FT IR) and Thermal Gravimetric Analysis (TGA).

I Ecoﬂex-MTPS graft copolymers were further characterized by Differential

Scanning Calorimetry (DSC) measurements.

0 Mechanical properties of ﬁlms made from Ecoﬂex-MTPS graft copolymers were

determined and compared to LDPE.

5.2 CONCLUSIONS

This section outlines the speciﬁc conclusions derived from the work in each of the

sections detailed above.

5.2.1 Extrusion polymerization of l, 4-dioxan-2-one (PDX) monomer and co-

polymerization with e-caprolactone (CL) monomer.

0 A twin-screw extrusion process to polymerize PDX monomer into high molecular
weight three-arm PPDX polymer was designed and engineered; three-arm PPDX
polymers of theoretical number-average molecular weight ranging from 50,000 to

100,000 g/mol were made with good control over molecular weight.

212

A polymerization temperature of 130°C was considered optimal on account of the
monomer-polymer thermodynamic equilibrium leading to depolymerization

above the melting temperature of the polymer.

An extruder screw conﬁguration comprising entirely of conveying elements was

found to be suitable for the extrusion polymerization of PDX monomer.

Maximum attainable monomer conversion in the case of PDX polymerization was

limited to 78%, close to the equilibrium monomer conversion value.

For molecular weights of 54,800 g/mol and 63,400 g/mol, conversion values are

identical even as monomer ﬂow rate increases from 23.23 to 40.75 g/min.

At a screw speed of 130 RPM for a PPDX molecular weight of 63,400 g/mol, as
temperature increases from 130 to 170°C, monomer conversion drops from 68%

to 61%.

Incorporation of 8, 11 and 16-mole % of CL monomer during the PDX
polymerization process made the conversion of PDX monomer reach almost

100%.

213

Lower intrinsic viscosity values of the PCL-co-PPDX copolymers as compared to
PPDX polymers of the same molecular weight could be attributed to a change in

conformation of the copolymers in solution.

Copolymer with 11-mol% CL is characterized by a melting temperature of 94°C,
close to the homopolymer melting temperature of 110°C. The melting temperature
of the co-polymers further reduces to 85°C and 68°C with incorporation of 16-
mol% and 24-mol% CL respectively. The glass transition temperature of the co-

polymer (1 l-mol% CL) decreases from -11°C (homo PPDX) to -22°C.

Thermal stability of the copolymers is substantially improved as compared to the

homopolymer.

Residence Time Distribution (RTD) studies indicated that the Axial Dispersed
Plug Flow Reactor Model provided a good ﬁt to the observed RTD and

conversion data.

Maximum dispersion (from Peclet number calculations) is obtained under
conditions of 130°C, 125 RPM and 22-ml/min monomer feed rate while least
dispersion is obtained at 170°C, 125 RPM and l4-ml/min monomer feed rate. At
130°C, due to the presence of low molecular weight pre-polymers along-with the
polymer melt, dispersion effects become pronounced. Dispersion effects increase

as overall throughput increases (increase of monomer feed rate).

214

 

5.2.2

Reactive extrusion cross-linking of biodegradable polyesters (Ecoflex and

PCL) and blending with talc.

Reactive Extrusion of micro-gelled cross-linking of PCL and Ecoﬂex was
accomplished in a twin-screw extruder using 2, 5-Bis (tert-butylperoxy)-2, 5-

dimethylhexane (Luperox 101) free radical initiator.

Samples removed from the vent port in the extruder exhibited lower values of gel
fraction as compared to those from the die proving that cross-linking chemistry is

dependent on the residence time in the extruder.

Increase in Luperox 101 concentration leads to an increase in gel fraction. PCL
polyester exhibits higher values of gel fraction as compared to Ecoﬂex for the

same Luperox 101 concentration.

Percent crystallinity and melting temperature reduce on cross-linking, as the size
and perfectness of cross-linked crystals are relatively low. Presence of impurities
such as catalyst residues, which act as nucleating agents, enhance the

crystallization process of cross-linked polyesters.

Rate of degradation of cross-linked polyesters is higher than regular polyester,

which is attributed to backbiting reactions in the presence of radicals.

215

Gel particles, as evinced by ESEM image, are of the order of 1-5 microns and are
termed as “rnicrogels”. These rnicrogels are uniformly distributed in the polyester

matrix and provide reinforcement.

Extrusion processing of Ecoﬂex with talc was accomplished in a twin-screw

extruder at a temperature of around 170°C.

Screw conﬁguration comprising conveying and kneading elements was found to
be effective in efﬁcient breakdown and dispersion of talc particles in the polyester

matrix.

Percent crystallinity and melting temperature reduce on blending with talc, as the
size and perfectness of the polyester crystals on blending are relatively low.
Presence of impurities such as catalyst residues, which act as nucleating agents,

enhance the crystallization process of polyesters in talc ﬁlled blends.

Rate of degradation of Ecoﬂex/talc and LEcoﬂex/talc blends are very similar to

pure Ecoflex and LEcoﬂex respectively.

Size of the talc particles, as evinced from ESEM images, are of the order of 1-10

microns and are uniformly distributed within the polyester matrix.

216

Tensile strength and break elongation values of talc ﬁlled polyesters are lower as

compared to the unﬁlled ones. Modulus increases on ﬁller addition.

0 Cross-linking increases the tensile strength and modulus (increase in molecular

weight), reducing the break elongation value.

I Tear and Puncture properties of Ecoflex/talc blends are better or comparable to

LDPE.

5.2.3 Ecoflex-Starch graft copolymers by reactive extrusion processing

0 Ecoﬂex-TPS graft copolymers were produced by reactive extrusion processing

using maleic anhydride or maleic acid as a trans-esteriﬁcation catalyst.

0 The graft copolymer was completely extractable in Dichloromethane using a
Soxhlet extraction technique. On solvent casting the extracted polymer solution, a
transparent ﬁlm was obtained. In the absence of maleic acid, only the polyester
was extracted out. These results conﬁrm that Ecoﬂex and TPS are covalently

linked, as opposed to a simple physical blend.

0 FTIR results of Ecoﬂex-TPS graft copolymer, showing the carbonyl stretch at
1720 cm", ester C-O stretch at 1270 cm" and the —OH polymer stretch at 3200-

3400 cm'1 conﬁrm the true existence of a graft copolymer.

217

Ecoflex-TPS graft copolymer exhibit degradation corresponding to both starch
and Ecoﬂex as opposed to simple physical blends, which show degradation

corresponding to only the polyester.

Tensile strength and modulus of elasticity of the graft copolymer are lower as
compared to LDPE. However, the graft copolymer exhibits break elongation

values of around 500%, which is much higher than LDPE.

Regular granular corn starch was maleated in a twin-screw co—rotating extruder
using maleic anhydride or maleic acid modiﬁers in the presence of glycerol as the

plasticizer and optional free-radical initiator such as Luperox 101.

Due to the low viscosity of MTPS, reactive maleation can be done at much higher

throughputs as compared to the regular thermoplastic starch (TPS) production.

The reactive maleation process prevents problems during TPS production such as

clogging of the thermoplastic starch melt and foaming at the die.

Carbonyl stretch peak at 1720 cm", observed for the extracted MTPS sample

(made using both maleic anhydride and maleic acid modiﬁer), conﬁrms reactivity.

Differential Scanning Calorimetry (DSC) results of MT PS samples exhibit

melting endotherms, not present in regular granular starch.

218

Thermal Gravimetric Analysis (TGA) results show that MTPS is more stable as
compared to either starch or TPS. At a temperature of 350.38°C, while 75% and
63% of starch and TPS have degraded respectively, only 58% of MTPS has

degraded.

Intrinsic viscosity of MT PS drops to 0.27 dllg as compared to 1.6 dllg and 1.2 dllg
for starch and TPS respectively. This reduction is attributed to a decrease in'
molecular weight due to the hydrolysis of starch in the presence of maleic acid.
However, in the presence of Luperox 101, free-radical initiator, intrinsic viscosity

increases to 0.45 dllg, which is attributed to cross-linking reactions taking place.

Ecoﬂex-MTPS graft copolymers were produced by the reactive extrusion

processing of Ecoﬂex and MTPS.

Ecoﬂex/MTPS (70/30 and 60/40; w/w) graft copolymers was completely
extractable in Dichloromethane using a Soxhlet extraction technique. On solvent
casting the extracted polymer solution, a transparent ﬁlm was obtained. This
reaction leading to the formation of a graft copolymer is dependent on the relative
amounts of Ecoﬂex and MTPS present (Does not hold true for more than 50%
MTPS). Cross-linked Ecoﬂex/MTPS (70/30;w/w) exhibits certain regions, which

are impermeable to chemical reaction.

219

FT IR results of Ecoflex/MTPS (70/30; w/w) graft copolymers (solvent cast ﬁlms
after extraction in Dichloromethane) made using both maleic anhydride and
maleic acid, show the carbonyl stretch at 1720 cm", ester C-O stretch at 1270 cm’
1 and the —OH polymer stretch at 3200-3400 cm'1 conﬁrming the true existence of

a graft copolymer.

DSC Results shows that while the melting temperature of the Ecoﬂex/MTPS
(70/30; w/w) graft copolymer is similar to pure Ecoﬂex, the crystallinity of the

graft copolymer is reduced.

TGA Results show that Ecoflex/MTPS (70/30; wlw) graft copolymer exhibits two
degradations, one corresponding to Ecoﬂex and the other corresponding to MTPS.

This conﬁrms the true existence of a graft copolymer.

Melt Pressures, during blown ﬁlm processing, decrease with increasing MTPS
fraction in the graft copolymer. However, for the same starch content in the graft
copolymer, Ecoﬂex/MTPS graft copolymer exhibits higher pressures as compared
to the Ecoflex/TPS graft copolymer. Moreover, for the same MTPS fraction, melt

pressures increase with the use of cross-linked Ecoﬂex.
Tensile strength and modulus of elasticity values of Ecoflex/MTPS graft

copolymers are comparable to LDPE and higher than Ecoﬂex/TPS graft

copolymer. The values increase and are better than LDPE for cross-linked

220

Ecoﬂex-MTPS graft copolymers (LEcoflex/MTPS). Break elongation values for
Ecoﬂex/MTPS graft copolymers are higher than pure Ecoﬂex and

LEcoﬂex/MTPS graft copolymer.

Tear and Puncture properties of Ecoﬂex/MTPS (70/30; w/w) graft copolymer are

comparable to LDPE.

221

Chapter 6

 

RECOMMENDATIONS

 

6.1 RECONIMENDATIONS

The recommendations with respect to the future work are explained in the following

sections.

6.1.1 Bulk Extrusion Polymerization of 1, 4-dioxan-2-one monomer

0 Detailed molecular weight analysis of Poly (1, 4-dioxan-2-one) (PPDX) and Poly
(e-caprolactone)-co-Poly (l, 4-dioxan-2-one) (PCL-co-PPDX) polymers produced

by reactive extrusion polymerization.

222

6.1.2

Mechanical properties of ﬁlms made from PPDX and PCL-co-PPDX polymers.
Dart, tear and puncture properties of ﬁlms made from PPDX and PCL-co-PPDX
polymers.

Economics of extrusion polymerization of PDX monomer and co-polymerization
of PDX monomer with CL monomer.

Extension of Axial Dispersed PFR model to polymerization of PDX monomer

and copolymerization of PDX monomer with CL monomer.

Reactive Extrusion Cross-linking of Biodegradable Polyester and Blending

with Inorganic Fillers.

Further investigation into cross-linking of PCL and Ecoﬂex using varying
amounts of Luperox 101 initiator.

Rheological studies on cross-linked Ecoflex (LEcoﬂex) and cross-linked PCL
(LPCL).

Mechanical Properties of Blends of Ecoﬂex and talc containing varying amounts
of talc (greater than 30-wt %).

Use of compatibilizers such as maleated Ecoﬂex (added externally or generated
insitu during extrusion processing) in Ecoﬂex/talc blends to improve mechanical
properties.

Water vapor and oxygen permeability studies on blends of Ecoﬂex/talc.

223

6.1.3 Polyester-Starch graft copolymers by Reactive Extrusion Processing

0 Incorporation of varying amounts of TPS content in the graft copolymers and
Study the effect on mechanical properties.

0 Further investigation into using other catalyst systems such as FASCAT and
Stannous Octoate instead of maleic anhydride and maleic acid in the formation of
Ecoflex/TPS graft copolymers.

0 Study effect of changing catalyst concentration on processing, thermal and
mechanical properties of graft copolymers.

0 Develop a method to quantify the exact amount of maleic anhydride and maleic
acid grafted onto the starch backbone.

I Rheological Studies on Maleated Thermoplastic Starch samples using various
amounts of modiﬁers.

0 Particle size distribution studies on starch, TPS and MTPS samples.

0 Mechanical, dart, tear and puncture properties on Ecoflex/MTPS graft
copolymers containing varying amounts of MTPS.

0 Detailed analysis on phase reversal of graft copolymers containing greater than
50% MTPS by conducting FTIR analysis.

0 Develop a single-step procedure for the production of Ecoﬂeme or
LEcoﬂex/MTPS graft copolymers. Compare mechanical properties with graft
copolymers formed by a two-step process.

0 Economics of Polyester-Starch graft copolymer production.

224

 

REFERENCES

1. S. Bloembergen, J. David, D. Geyer, A. Gustafson, J.Snook, and R. Narayan,
Biodegradﬁable Plastics rmd Polymers, 601, Elsevier, New York (1994).

2. R. Narayan, Paradigm for Successful Utilization of Renewable Resources, 78,
AOCS Press, Illinois (1998).

3. R. Narayan, American.Chem.Soc.Syﬂ2., 2, 575 (1994).

4. R. Narayan, Biodeﬂdable Pla_stics and Polymers, 261, Elsevier, New York
(1994).

5. H. Inata and S. Matsumara, Journal of Applied Polymer Science, 32, 1877 (1995).
6. U.S.Patent 4,128,599 (1978).

7. D. Carlson, L. Nie, R. Narayan, and P. Dubois, Journal of Applied Polymer
Science, 72, 477 (1999).

 

8. D. Bikaris, and C. Panayiotou, Joumal of Applied Polymer Science, 70, 1503
(1998).

9. UR. Vaidya and M. Bhattacharya, Journal of Applied Polymer Science, 52, 617
(1994)

10. C. Maier and M. Lambla, PolymerEngineeringgnd Science. 35 (1995).
11. US. Patent 5,310,945 (1994).

12. US. Patent 2,803,646 (1966).

13. Y.M. Cao and D.M. Yu, Mater.Chem.Eng., 3, 1921 (1996).

14. M.A. Sabino, S. Gonzalez, L. Marquez and LL. Feijoo, Polmer Deggdation and
Stability, 69, 209 (2000).

15. M.A. Sabino, L. Sabater, G. Ronca and A.J. Muller, Polmer Bulletin, 48, 291
(2002).

16. K. Yang, X. Wang and Y. Wang, Journal of Macromolecular Science Part C-
Polymer Reviews, 3, 373 (2002).

17. H. Nishida, M. Konno and Y. Tokiwa, Polymer Degadationgd Stability. 68,
271 (2000)

225

 

18. KM. Stridsberg, M. Ryner, A. Albertsson, Advances in Polymer Science, 157, 41
(2002).

19. RF. Storey and J .W. Sherman, _M_acromolecules, 35, 1504 (2002).
20. A. Duda and S. Penczek, Macromolecules, 28, 5981 (1995).
21. A. Rehor, N. Tirelli and J .A. Hubbel, Macromolecules, 35, 8688 (2002).

22. JP. Claverie, S. Viala, V. Maurel and N. Vincent, Macromolecules, 34, 382
(2001).

23. M. Gadzinowski, S. Sosnowski and S. Slomkowski, mcromolecules, 29, 6414
(1996).

24. J. Raquez, P. Degée, R. Narayan and P. Dubois, Macromol.Rapid Commun., 21,
1063 (2000).

25. US. Patent 3,021,309 through 3,021,317 (1962).
26. US. Patent 3,063,967 (1962).

27. US. Patent 3,063,968 (1962).

28. US. Patent 3,391,126 (1968).

29. US. Patent 3,645,945 (1972).

30. GB. Patent 1,540,053 (1979).

31. US. Patent 5,225,520 (1993).

32. US. Patent 5,100,433 (1992).

33. US. Patent 4,559,945 (1985).

34. US. Patent 4,441,496 (1984).

35. US. Patent 4,653,497 (1987).

36. HR. Kricheldorf and D. Damrau, Macromol.Chem.Phys.. 199, 1089 (1998).

37. H. Nishida, M. Yamashita, T. Endo and Y. Tokiwa, Macromolecules, 33, 6982
(2000).

226

38. H. Nishida, M. Yamashita, M. Nagashima, T. Endo and Y. Tokiwa, Journal of
Polymer Science: Part A: Polymer Chemistry, 38, 1560 (2000).

39. J. Raquez, P. Degée, R. Narayan and P. Dubois, Macromolecules, 34, 8419
(2001).

40. M. Krishnan, PhD Thesis, Michigan State University (1998).
41. H. Hooker, J .M.S.-Pure Ale.Chem., A30, 595 (1993).
42. H. Jacobson and W.H. Stockmayer, J .Chem.Phys., 18, 1600 (1950).

43. A. Duda, T. Biela, J. Libiszowski, S. Penczek, P. Dubois, D. Mecerreyes and R.
Jerome, Polymer DeggadatioL and Stability, 59, 215 (1998).

44. A. Duda and S. Penczek, Macromolecules, 23, 1636 (1990).

45. A. Evstropov, B.V. Lebdev, E.G. Kiparisova, V.A. Alekseev and GA. Stashina,
Vysokomol Soedin, 22, 2450 (1980).

46. A. Albertsson and I.K. Varma, Advances in Polymer Science, 157, 1 (2002).

47. S. Penczek, P. Kubisa and K. Matyjaszewski, Advances in Polymer Science:
Cationic Ring-Omning Polmerization Part H (Synthetic Applications), Spinger-
Verlag- New York.

48. U.S. Patent 3,597,450 (1971).

49. H. Nishida, M. Yamashita, M. Nagashima, N. Hattori, T. Endo and Y. Tokiwa,
Macromolecules, 33, 6595 (2000).

50. A. Kowalski, A. Duda and S. Penczek, Macromol.Rapid Commun., 24, 2266
(1998).

51. A. Schindler, Y.M. Hibionada and CG. Pitt, J.Polyrpn.Sci Polm.Chem.Ed., 20,
319 (1982).

52. KS. Bisht, Y.Y. Svirkin, R.A. Gross, D.L. Kaplan and G. Swift, @cromolecules,
29, 7759 (1996).

53. K. Ishikiriyama, M. Pyda, G. Zhang, T. Forschner, J. Grebowicz and B.
Wunderlich, J .Macromol. Sci. Phys, B37, 27 (1998).

54. R. J. Hernandez, S.E.M. Selke and JD. Culter, Plastics Packaging-Mommies,
Processing, Applications and Regplations, Hanser (2000).

227

55. D. Carlson, L. Nie, R. Narayan and P. Dubois, JournJal of Applied Polmer
Science, 72, 477 (1999).

 

56. C. Maier and M. Lambla, Polymer Erpg'meering and Science. 35, 1197 (1995).
57. C. Maier, M. Lambla, D. Bigio, Proceedings of 52 SPE Antec, 40, 123 (1994).

58. X. Wang, C. Tzoganakis and G.L. Rempel, Journal of Applied Polymer Science,
61, 1395 (1996).

59. H. Inata and S. Matsumara, Journal of Applied Polymer Science, 32, 1877 (1995).
60. U.S. Patent 4,128,599 (1978).

61. D. Bikiaris and C. Panayiotou, Journal of Applied Polymer Science, 70, 1503
(1998).

62. Z. Yang, M. Bhattacharya, Polmer, 37, 2137 (1996).
63. C. Rauwendaal, Polymer Extrusion, Hanser (2001).
64. U.S. Patent 5, 801, 224 (1998).

65. W. Michaeli, H. HOcker, U. Berghaus and W. Frings, Journal of Applied Polymer
Science, 48, 871 (1993).

 

66. H. Kye and J .L. White, Journal of Applied Polymer Science, 52, 1249 (1994).
67. U.S. Patent 5,412,005 (1995).

68. U.S. Patent 3,642,964 (1972).

69. German Offenlegungsschrift 2,302,564 (1974).

70. German Offenlegungsschrift 2,423,764 (1975).

71. U.S. Patent 4,105,637 (1978).

72. U.S. Patent 4,115,369 (1978).

73. U.S. Patent 4,073,773 (1978).

74. U.S. Patent 4,443,592 (1984).

228

75. J. Gimenez, M. Boudris, Ph. Cassagnau, A. Michel, Polymer Reaction
Engineering, 8, 135 (2000).

76. J. Gimenez, M. Boudris, Ph. Cassagnau, A. Michel, International Polymer
Processing, 15, 20 (2000).

77. J. Gimenez, Ph.D Thesis, University of Claude Bernard Lyon 1 (1999).

78. A. Poulesquen, B. Vergnes, Ph. Cassagnau, J. Gimenez and A. Michel,
Intemationﬂolmer Processing, 16, 31 (2001).

79. U.S. Patent 5,801,224 (1998).

80. U.S. Patent 5,906,783 (1999).

81. U.S. Patent 5,969,089 (1999).

82. Young, ACS Symmsium Ser., 152 (1977).

83. U.S. Patent 5,578,691 (1996).

84. W. Thiele, P_lastics Formulating& Compounding, 2, 6 (1996).

85. HS. Fogler, Elements of Chemical Reaction Epgineering 3rd Edition, Prentice
Hall (1999).

86. M. Thompson, J.P. Puaux, A.N. Hrymak and A.E. Harnielec, International
Polymer Processing, 10, 111 (1995).

87. J .D. Dziezak, Food Technology, 44, 163 (1989).
88. B. Singh and SH. Rizvi, Journal of Food Process Engineering, 21, 91 (1998).
89. S. Middleman, Fundamentals of Polmer Processing, McGraw-Hill (1977).

90. B. Siadat, M. Malone and S. Middleman, Polymer Engineering and Science, 19,
787 (1979).

91. D. Bounie, Journal of Food Engineering, 7, 223 (1988).

92. R. Giudici, C.A. Oller, D.N. Isabel, C. Beiler and N. Scherbakoff, Journal of
Applied Polmer Science, 67, 1573 (1998).

93. W. Michaeli, A. Grefenstein, Polymer Engineeringand Science, 35, 1485 (1995).

229

94. M. Zhang and W.H. Ray, Journal of Applied Polymer Science, 86, 1047 (2002).

95. S. Hojabr, W.E. Baker, K.E. Russell, P]. McLellan and M.A. Huneault,
International Polmer Processing, 13, 118 (1998).

96. AI. Yeh and Y.M. Jaw, Journal of Food Epgineering, 39, 81 (1999).

 

97. V. Komolprasert and R.Y. Ofoli, Biotechnology and Bioengineering, 37, 681
(1991).

98. J. Raquez, MS. Thesis, University of Mons-Hainaut (1999).

99. S. Balakrishnan, M. Krishnan, P. Dubois and R. Narayan, Polymer Engineering
and Science (In Press).

100.Kim, W.S. Kim, D.H. Lee, I.R. Jeon and K.H. Seo, Journal of Applied Polymer
Science, 81, 1115 (2001).

 

101. DJ. Kim, H.J. Kang and K.H. Seo, Journal of Applied Polymer Science, 81, 637
(2001).

102. Y.K. Han, P.G. Edelman, 8.]. Huang, Journal of Macromolecular Science—Part A
Pure and Applied Chemistpy, 25, 847 (1988).

103. D. Pramanick and T. Ray, T. Polym Bull, 19, 365 (1988).
104. D. Pramanick and T. Ray, T. Biomateriaﬁ, 8, 407 (1987).
105. http://www.campoly.com/notes/005.pdf.

106. S.H. Katz, HandbooJk of Filleraand Reinforcementa for Pla_stics, Van Nostrand
Reinhold Co. (1978).

107. CD. Han, Journal of Applied Polymer Science, 18, 821 (1974).

108. G.V. Vinogradov, A.YA. Malkin, E.P. Plotnikova, O.Yu. Sabsai and NE.
Nikolayeva, International Journal of Polymer Materiala, 2, 1 (1972).

109. RH. Beck Jr., SPE Annual Technical Conference, 29, 392 (1971).

110. RC. Hall, C.G. Cashulette, J.A. Valentine and GD. Chadderon, SAMPE
Technical Conference, 7 (1975).

111. RA. Willermet, R.A. Pett and T.J. Whalen, American Ceramic Society Annual
Conference (1974).

230

112. R]. Farris, Transactioms of Society of Rheology, 12, 281 (1968).

 

113. AA. Berlin, S.A. Volfson, N.S. Enikolopian and SS. Negmatov, Principles of
Polmer Composites, Springer-Verlag (1985).

114. F. Rodriguez, Principles of Polymer Systems 2nd Edition, McGraw Hill (1982).

l 15. http://www.dow.com/webapps/lit/litorder.asp?ﬁ @athﬂoneﬁpde/noregBZ l-
00055.pdf&pdf=true.

116. D. Carlson, P. Dubois, L. Nie and R. Narayan, Polymer Engineering and
Science, 38, 311 (1998).

 

117. H. Humki, ﬂastics Additives, 3rd Edition, Hanser; New York, 22 (1990).
118. B. Plage, H. Schulten, Momolecules, 23, 2642 (1990).

119. J .L. Bolland, G. Gee, Trans Faraday Soc, 42, 236 (1946).

120. J .L. Bolland, Trans Faraday Soc, 44, 669 (1948).

121. L.E. Nielsen, Mechanical Properties of Polymers. Reinhold; New York, 115
(1962).

122. R]. Flory, Journal of American Chemical Society, 67 , 2048 (1945).
123. http://www.opta-food.com/access/starchg.html.
124. DJ. Manners, Carbohydrate Polymers, ll, 87 (1989).

125. W.R. Morrison, R.F. Tester, M.J. Gidley and J. Karkalas, Carbohydrate
Research, 245, 289 (1993).

126. HP. Zobel, Starch, 40, 44 (1988).

127. J. J.G. Van Soest, S.H.D. Hulleman, D.de. Wit, J .F.G. Vliiegenthart, Industrial
Crops and Products, 5, 11 (1996).

128. I.D. Evans, D.R. Haisman, Starch, 34, 224 (1982).

 

129. J .W. Donovan, Biopolymers, 18, 263 (1979).

130. J.J.G. Van Soest, D.de. Wit, H. Toumois, J.F.G. Vliliegenthart, Carbohydrate
Research (1995).

231

 

131.B.Y. Shin, S. Lee, Y.S. Shin, S. Balakrishnan and R. Narayan, Polymer
Engineeringand Science (In Press).

132. G.J.L. Grifﬁn, Chemistry and Technology of Biodegradable Polymers, Blackie
Academic and Professional (1992).

133. H. Pranarnuda, Y. Tokiwa and H. Tanaka, Journal of Environmental Polymer
Degradation, 4,1 (1996).

134.B.A. Ramsay, V. Langlade, P.J. Carreau, and J.A. Ramsay, Applied and
Environmentai Microbiology, 59, 1242 (1993).

135. W. Wiedmann, and E. Strobe], Starch, 43, 138 (1991).
136. R.L. Shorgen, G.F. Fanta, and W.M. Doan, Starch, 45, 276 (1993).

137. P. Forssell, J. Mikkila, and T. Sourtti, J.M.S. Pure Appl. Chem., A33, 703
(1996).

138. R. Narayan, Polymers from Agr_icultural Co products in ACS Symposmm Ser.,
2, 575 (1994).

139. J .J .G. Van Soest, K. Benes, and D.de. Wit, Polymer, 37, 3543 (1996).

140. A. Rudin, The Elements of Polymer Science and Epgineering, Academic Press:
New York (1982).

141. W.J. MacKnight, F.E. Karasz and J .R. Fried, Polymer Blends: Vol 1, Academic
Press, 185 (1978).

142. DR. Paul, Polymer Blends-Vol 1 and 2, Academic Press (1978).

143.1. Ghosh, Master Thesis, Virginia Polytechnic Institute and State University
(1998).

144. US Patent 5,462,983 (1985).
145. D. C. Argyropoulus, Master Thesis, Michigan State University (1992).
146. J .B. Snook, Master Thesis, Michigan State University (1994).

147. R. Fayt, R. Jerome and P. Teyssie, Polmrer Engineering and Science, 24, 25
(1986).

232

 

148. R. Fayt, R. Jerome and P. Teyssie, Polymer Epgineering and Science, 27, 328
(1987).

149. GR Fanta and EB. Bagley, Encyclopedigif Polymer Science, John Wiley &
Sons; New York (1970).

150. GP. Fanta, Block and Graft Copolymers- Vol I, John Wiley & Sons; New
York (1973).

151. RH. Otey, R.P. Westhoff and W.M. Doane, Industrial Engineering Chemistpy
Products Research Development, 19, 592 (1980).

152. RH. Otey and RP. Westhoff, Industrial Epﬂreering Chemist_ry Products
Research Development, 23, 284 (1984).

153. RH. Otey, R.P. Westhoff and W.M. Doane, Industrial Engineering Chemistry
Products Research Development, 26, 1659 (1987).

154. US Patent 5,540,929 (1996).

155. R.L. Whistler, Advanced Carbohydrate Chemistry (1945).

156. A.M. Mark and CL. Mehltretter, Die Starke, 24 (1972).

157. L.H. Kruger and M.W. Rutenberg, Derivatives of stamp (1984).
158. U.S. Patent 2,461,139 (1949).

159. U.S. Patent 2,661,349 (1953).

160. P. Tomasik, Y.J. Wang and J. Jane, Starch, 47, 96 (1995).

233

 

    

 
 

11111111111

RRRRR

111111111

429