Mechano Optical Behavior of Novel Polymers for Capacitor

MECHANO OPTICAL BEHAVIOR OF NOVEL POLYMERS FOR CAPACITOR

APPLICATION DURING THEIR PROCESSING CYCLES

A Dissertation Presented to The Graduate Faculty of the University of Akron

In Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy

Ido Offenbach September, 2016

MECHANO OPTICAL BEHAVIOR OF NOVEL POLYMERS FOR CAPACITOR

APPLICATION DURING THEIR PROCESSING CYCLES

Ido Offenbach Dissertation

Approved: Accepted: ______Advisor Department Chair Dr. Mukerrem Cakmak Dr. Sadhan Jana

______Co-Advisor/Committee Member Dean of the College Dr. Robert A. Weiss Dr. Eric Amis

______Committee Member Interim Dean of the Graduate School Dr. Mark Soucek Chand Midha

______Committee Member Date Dr. Abraham Joy

______Committee Member Dr. Chrys Wesdemiotis

ABSTRACT

This work is a part of collaborative project between Multidisciplinary University

Research Initiative (MURI) through which an advanced polymeric capacitor films for military applications were designed. Two of those novel polymers were PPOH (hydroxyl functionalized isotactic polypropylene with comonomer of 10-hydroxy-1-undecen) and

PI(BTDA-DAH) (Polyimide 3,3',4,4'-benzophenone tetracarboxylic dianhydride and 1,6- diaminohexan). This dissertation focused on the effect of processing conditions on the mechano-optical behavior of PPOH and PI(BTDA-DAH).

Firstly, the real-time mechano-optical behavior of PPOH containing 0.4 mol % comonomer and its comparison with unmodified polypropylene (PP) were studied in the partially molten state during processing cycle of heating, stretching, annealing, and cooling. It was revealed that the crystalline network dominated the material response during the processing cycle for both polymers. However, the presence of hydrogen bonding between the hydroxyl groups in PPOH was found to affect the structural evolution of the

PPOH copolymer significantly more than compared to the PP homopolymer.

Secondly, the real-time mechano-optical behavior of PI(BTDA-DAH) was studied in the glassy and the rubbery states as a function of processing temperature and stretching rate during uniaxial deformation. Thee regimes of stress optical behavior were revealed.

First, at the early stage of deformation the stress optical rule is observed; birefringence linearly increased with a stress optical constant of 17.8 GPa-1 - regime I. Second, a

iii

deviation from linearity took place. At higher temperature and/or lower stretching rate the deviation is positive and the birefringence rapidly increases while the stress slowly increases- regime II. At lower temperature and/or higher stretching rate this deviation of linearity is negative- regime IIIa. Third, in cases where regime II is revealed, it was followed by a negative deviation of birefringence and reached a plateau while stress rapidly increased -regime IIIc. According to off-line characterization techniques: differential scanning calorimetry and wide-angle X-ray diffraction showed that the material remains amorphous during regime I and the early stage of regime II. By the end of regime II, a rapid increase in the crystallinity was observed. This implies stress induced crystallization associated with regime II. There was no significant change in the crystallinity with further stretching into regime III where the polymer chains reach to their finite extensibilities.

Thirdly, the real-time infrared-mechano-optical behavior of PI(BTDA-DAH) during uniaxial deformation followed by relaxation was studied. The relaxation showed high dependency on the polymer structure pre-relaxation. Three regimes of stress optical behavior and stress reho (orientation function) behavior were revealed. At the early stage of deformation (regime I), birefringence and orientation function linearly decreased with a stress. At early stage of deformation in regime II, the birefringence and orientation function slowly decrease. These behaviors changed with the deformation level in regime II. At intermediate stage of the deformation in regime II, the birefringence and orientation function did not change while the stress rapidly decreases, and then birefringence and orientation function started slowly to increase, while the stress slowly decreased. At the high stage of the deformation in regime II, the birefringence and orientation function were slowly increased while the stress rapidly decreases, and then birefringence and orientation

function rapidly increased while the stress slowly decreases. At the high stage of the deformation, regime III, the birefringence and orientation function were slowly increased while the stress rapidly decreased, and then birefringence slowly increased while the stress slowly decreased. It also was found that the birefringence and orientation function increased with enough relaxation time at all level of deformation in regime II.

Fourthly, the real-time mechano-optical behavior of PI(BTDA-DAH) during biaxial deformation and their effect on dielectric properties was studied. It was found that during the biaxial stretching, the phenyl groups in the PI(BTDA-DAH) chains became parallel to the surface plane which reduced the polarizability of the polymer chain in the film thickness direction. As a result, the dielectric constant decreased with increasing in the stretching ratios.

DEDICATION

This dissertation is dedicated to my parents Zeev and Ester Offenbach, who made me believe that the sky is the limit and encouraged me to never give up on dreams.

ACKNOWLEDGEMENTS

I would like to thank my academic advisor Dr. Mukerrem Cakmak for his continuous guidance, encouragement and trust.

I am equally indebted to Dr. Robert A. Weiss for continuously motivating me throughout my research.

I would also like to thank my dissertation committee members: Dr. Chyrs

Wesdemiotis, Dr. Abraham Joy, and Dr. Mark Soucek for their guidance and taking time- off from their busy schedules.

I would like to acknowledge our project collaborators, Dr. T.C. Mike Chung at the

Pennsylvania State University and Dr. Gerege Sotzing, Dr. Rampi Ramprasad, Dr. Yang

Cao, at the University of Connecticut, for useful discussions on several topics and highly productive collaboration.

Thank you to Dr. Xuepei Yuan for synthesizing the hydroxyl-modified polypropylene copolymers; to Dr. Rui Aa and Gerege Terrich for synthesizing the

PI(BTDA-DAH); and to JoAnne Ronzelle, Mattewos Teffri and Zongze Li for performing the dielectric measurements.

Special thanks to Dr. Sahil Gupta for his ongoing advice, guidance and friendship.

Thank you to my colleagues in the College of Polymer Science and Engineering for their support and assistance.

vii

I am grateful for the opportunity to have worked for the Office of Naval Research

(ONR) on a Multidisciplinary University Research Initiative (MURI) grant

(Contract No. N00014-10-1-0944).

I am so very thankful to Wilhite family for their support, help and friendship in the last four years.

I would like to thank my brother and my sister for always being there for me and for good advice.

My wife, Effie, for her unconditional love and support. And to Pitzie, who came into our lives and changed our family.

Thank you to my grandparents for all our good times and for all the memories that we share.

And a big thank you to the Schuldiner family being my home away from home; especially to Dr. Ruth Schuldiner-Rosaler and Dr. Michael Schuldiner for their dedication to this dissertation.

viii

Table of Contents

LIST OF TABLES ...... xiv

LIST OF FIGURES ...... xv

KEYWORDS ...... xxix

LIST OF ABBREVIATIONS ...... xxx

CHAPTER I ...... 1

INTRODUCTION ...... 1

LITERATURE REVIEW ...... 3

2.1 Energy Storage Systems ...... 3

2.2 Types of Energy Storage Systems ...... 4

2.2.1 Battery ...... 4

2.2.2 Fuel Cell ...... 5

2.2.3 Superconducting magnetic energy storage (SMES) ...... 5

2.2.4 Flywhell ...... 6

2.2.5 Capacitors ...... 6

2.3 Interaction of the dielectric constant and Light...... 9

2.3.1 Refractive index ...... 9

2.3.2 Birefringence ...... 10

2.3.3 Measurement anisotropy optical properties ...... 12

2.4 Polymer crystallization ...... 17

2.4.1 Fast crystallization polymer ...... 18

2.4.2 Slow crystallization polymer ...... 18

2.4.3 Thermal crystallization ...... 19

2.4.4 Strain Induced Crystallization ...... 21

2.5 Polypropylene ...... 22

2.6 Polyimides ...... 23

2.7 Processing techniques for thin films in capacitor technology...... 25

2.7.1 Melt Cast Film By Extrusion ...... 25

2.7.2 Tenter-frame ...... 26

2.7.3 Blown Film ...... 28

2.7.4 Double Blown Film ...... 29

2.7.5 Solution Casting ...... 30

2.7.6 Processing of fast crystallization polymers in partial molten state...... 31

2.7.7 Processing of slow crystallization polymers between glass transition and cold

crystallization temperatures...... 34

2.8. FTIR Spectroscopy ...... 37

REAL-TIME INFRARED−MECHANO-OPTICAL BEHAVIOR AND THE

STRUCTURAL EVOLUTION OF POLYPROPYLENE AND HYDROXYL-

FUNCTIONALIZED POLYPROPYLENE DURING UNIAXIAL DEFORMATION 46

3.1 Introduction ...... 47

3.2 Experimental Details ...... 49

3.2.1 Materials ...... 49

3.2.2 Sample Preparation ...... 49

3.2.3 Characterization ...... 50

3.3 Results and Discussion ...... 54

3.3.1 Mechano-Optical Behavior ...... 55

3.4 Structural Evolution ...... 77

3.6 Conclusions ...... 79

REAL-TIME MECHANO-OPTICAL BEHAVIOR AND STRUCTURAL EVOLUTION

OF POLYIMIDE (BTDA-DAH) DURING UNIAXIAL DEFORMATION...... 80

4.1 INTRODUCTION ...... 81

4.2 Experimental Details ...... 83

4.2.1 Materials ...... 83

4.2.2 Sample Preparation ...... 84

4.2.3 Characterization ...... 85

4.3 Results and Discussion ...... 87

4.3.1 Mechanical Behaviors ...... 88

4.3.2 Mechano-Optical Behavior ...... 90

4.3.3 Strain Optical Behavior ...... 95

4.3.4 Structural Studies: Effect of Deformation...... 98

4.3.5 Structural Evolution ...... 104

4.4 Conclusions ...... 106

ROLE OF RELAXATION ON STRAIN INDUCED CRYSTALLIZATION OF

UNIAXIALLY STRETCHED PI(BTDA-HDA) FILMS: REAL-TIME INFRARED-

MECHANO-OPTICAL BEHAVIOR AND ELECTRICAL STUDY ...... 107

5.1 Introduction ...... 109

5.2 Experimental Details ...... 111

5.2.1 Materials ...... 111

5.2.2 Sample Preparation ...... 112

5.2.3 Characterization ...... 113

5.3 Results and Discussion ...... 116

5.4.1 Depression of melting point ...... 117

5.3.2 Mechanical Behavior ...... 118

5.3.3 Optical Behavior ...... 121

5.4.4 Rheo-Optical Behavior ...... 123

5.3.5 Mechano-Optical Behavior ...... 127

5.3.6 Mechano- Rheo-Optical Behavior ...... 129

5.3.7 Structural Studies: Effect of Stretching Ratio ...... 131

5.3.8 Structural Studies: Effect of Relaxation Time ...... 138

5.3.9 Structural Evolution ...... 143 xii

5.3.10 Dielectric properties ...... 145

5.4 Conclusions ...... 146

REAL-TIME MECHANO-OPTICAL BEHAVIOR AND STRUCTURAL EVOLUTION

OF PI (BTDA-DAH) DURING BIAXIAL DEFORMATION...... 148

6.1 Introduction ...... 149

6.2 Experimental Details ...... 150

6.2.1 Materials ...... 150

6.2.2 Sample Preparation ...... 151

6.2.3 Characterization ...... 152

6.3 Results and Discussion ...... 155

6.3.1 Mechanical Behaviors ...... 155

6.3.2 Optical Behaviors ...... 158

6.3.3 Strain Optical Behavior...... 164

6.3.4 Structural Evolution ...... 169

6.3.5 Dielectric properties...... 173

6.4 Conclusions ...... 177

6.5 Appendix ...... 179

CONCLUSION ...... 180

REFERENCES ...... 184

xiii

LIST OF TABLES

Table 3. 1. Processing conditions for preparing unoriented polypropylene films...... 50

Table 3. 2. Thermal properties of PP and PPOH polymers before and after each processing stage...... 63

Table 3. 3. Mechanical properties of PP and PPOH polymers...... 67

Table 4. 1. Thermal properties of PI(BTDA-DAH) polymer films after uniaxial stretching at different stretching ratios at a constant temperature of 105°C and strain rate of

0.0024sec-1...... 103

Table 4. 2. Thermal properties of PI(BTDA-DAH) polymer films after uniaxial stretching at varying temperatures and a strain rate of 0.0024sec-1...... 103

Table 6.1 Refractive indices of PI(BTDA-DAH) which are stretched to different stretching ratios during simultaneous and sequential biaxial stretching at 90°C and 50 mm/min. . 179

xiv

LIST OF FIGURES

Figure 2.1 Regone plot for different energy storage systems. [7] (adapted with permission from “Electropaedia ” http://www.mpoweruk.com/alternatives.htm.

Figure 2.2 Structure of sandwich capacitors. [11] (Adapted with permission from “The

Figure 2.3 ELECTOMAGNETIC VAVE. [13] (adapted with permission from ref.

“electromagnetic radiation” by m. abramowitz, m. j. parry-hill, and m. w. davidson, https://micro.magnet.fsu.edu/primer/java/scienceopticsu/electromagnetic/index.html). ... 9

Figure 2.4 Reflection and refraction of a ray. [14] Adapted from ref. [14] ...... 10

Figure 2.5 Propagation of light throughout the calcite crystal which is birefringent medium. [15] (adapted with permission from “Microscopy” by B. Murphy, K. R. Spring,

T. D. Fellers , and M. W. Davidson, http://nikon.magnet.fsu.edu/print/articles/polarized/birefringence-print.html )...... 11

Figure 2.6 Mechano-optical behavior of different type of polymer during uniaxial deformation. a deviation from SOR due to stretching temperature close to Tg. B.

Deviation from s.o.r due to high elongation chain. C. Divination from s.o.r due to crystallization. [23] (adapted with permission from “A birefringence study of polymer

crystallization in the process of elongation of films” by D. S. Ryu, T. Inoue, K. Osaki.

Figure 2.7 Mechano-optical behavior of PS which were stretched at different stretching temperature. [24] (adapted with permission from “Stress Birefringence in Amorphous

1977, John Wiley & Sons, Inc.)...... 16

Figure 2.8 Stress optical behavior of amorphous PET during uniaxial stretching. [23]

(adapted with permission from “A birefringence study of polymer crystallization in the process of elongation of films” by D. S. Ryu, T. Inoue, K. Osaki. Polymer 1998, 12,

Figure 2.9 Growth rate of Poly(tetramethyl-p-silphenylene)siloxane spherulite during crystallization as function of temperature and molecular weight. [32] (adapted with permission from “Crystallization of Poly‐(Tetramethyl‐p‐Silphenylene)‐Siloxane

Polymers” by J. H. Magill . Polymers. Journal of Applied Physics, 1964, 35, 3249-3259.

Figure 2.10 Syntesis of PI. [50] (adapted with permission from “Synthesis,

Characterizations, Aging and Semiconductor Device Passivation” by S. Diaham, M. L.

Locatelli, R. Khazaka, InTech Polymers. under CC BY 3.0 license. Available from: http://dx.doi.org/10.5772/53994)...... 24

Figure 2.11 Temperature profile offast crystallization polymer during bi-axial processing cycling by Tenter frame machine. (a) Tenter frame machine, (b) Temperature profile of

PP, and (c) Thermal behavior (DSC) of fast crystallization polymer...... 27

xvi

Figure 2.12 Temperature profile of slow crystallization polymer during bi-axial processing cycling by Tenter frame machine. (a) Tenter frame machine, (b) Temperature profile of PP, and (c) Thermal behavior (DSC) of slow crystallization polymer...... 28

Figure 2.13 schematic of blown film process. [56] (adapted with permission from

“Production of films, containers, and membranes“ by W. McKeen in Permeability

Properties of Plastics and Elastomers: A Guide to Packaging and Barrier Materials”.

Figure 2.14 schematic of double blown film process. [57] (Reprinted with permission from “Films” by T. Brink, http://extrusionist.com/extrusion-processes/films/index.html ).

...... 30

Figure 2.15 Mechano-optical behavior of PP during stretching at its partial molten state.

[58] (adapted with permission from “Real time development of structure in partially molten state stretching of PP as detected by spectral birefringence technique” by Y.

Ltd.)...... 32

Figure 2.16 Herman Orientation function of crystalline and amorphous regions during polyaxially stretching. the orientation function was measured in respect to the normal direction. [61] (adapted with permission from “Studies on the Stretching of

Polypropylene Film. II. Polyaxial Stretching” by S. Okajima, S. Homma. Journal of

Inc.)...... 33

Figure 2.17 True stress-birefringence plots for pla films stretched at (a) 60%/min at various temperatures. (b) 240%/min at various temperatures. (c) at 960%/min at various

xvii

temperatures. [27] (adapted with permission from “Nonlinear Mechanooptical Behavior of Uniaxially Stretched Poly(lactic acid: Dynamic Phase Behavior” by J. Mulligan, M.

Society)...... 35

Figure 2.18 Structural model representative of PENs behavior under rubbery state deformation. [28] (adapted with permission from “Large Deformation Mechano-Optical and Dynamical Phase Behavior in Uniaxially Stretched Poly(ethylene naphthalate)” by C.

American Chemical Society)...... 36

Figure 2.19 Michelson FTIR interferometer. [68] (adapted with permission from “Fourier

Transform Infrared Spectrometry” by P. R. Griffiths, J.A.De Hasetch, J. D. Winefordner.

Figure 2.20 Top view of URS-FTIR with the red laser beam travel path...... 41

Figure 2.21 Orientation function of PP during uniaxial stretching at 29°C. 풇풂- orientation function of amorphous regime, 풇풄- orientation function of crystalline regime, and 풇풂풗- average orientation function of crystalline and amorphous regimes. [75] (adapted with permission from “Infrared dichroism, molecular structure, and deformation mechanisms of isotactic polypropylene” by R. J. Samuels. Die Makromolekulare Chemie Supplement

Figure 2.22 Orientation function of crystalline region of PP during uniaxial stretching different stretching temperature. [76] (adapted with permission from “Deformation- induced morphology evolution during uniaxial stretching of isotactic polypropylene:

xviii

Figure 2.23 Orientation functions of PAA(BPDA/PDA:ODA) and PI(BPDA/PDA:ODA) with different compositions during uniaxial stretching. PDA:ODA ratios: (a) 70:30, (b)

50:50 and (c) 30:70. [77] (adapted with permission from Spontaneous Molecular

Orientation of Polyimides Induced by Thermal Imidization. 3. Component Chain

Orientation in Binary Polyimide Blends” by M. Hasegawa, K. Okuda, M. Horimoto, Y.

American Chemical Society)...... 43

Figure 2.24 Real time molecular mechanism during thermal imidization. [73] (adapted with permission from “Molecular mechanism of temporal physico/chemical changes that take place during imidization of polyamic acid: Coupled real-time rheo-optical and IR dichroism measurements” by E. Unsal, M. Cakmak. Polymer 2014, 55, 6569-6579.

Figure 3.1 Synthesis of PPOH polymer...... 49

Figure 3.2 DSC curves for the PP and PPOH polymers obtained from the undeformed films. The symbols indicate the following: Tm,onset – onset of melting, Ts – optimum stretching temperature determined from eq. (6), Tm – melting point., ∆Hf – heat of fusion,

m – mass fraction crystallinity of the polymer...... 54

Figure 3.3 True stress of PP and PPOH at Ts-31C, Ts and Ts+14C during four stages of processing cycle: (a) heating + 1 minute of holding , (b) stretching, (c) 30 min of annealing, and (d) cooling. Solid lines represent PP while the dash lines represent PPOH.

xix

The stress-strain curve in the 2nd stage can be divided into 2 regimes as shown for PP stretched at 110C, I: elastic flow, and II: yielding and plastic flow. The inset in (b) shows the elastic deformation region...... 57

Figure 3.4. In-plane birefringence of PP and PPOH at Ts-31C, Ts, and Ts+14C during four stages of processing cycle: (a) heating + 1 minute of holding, (b) stretching, (c) 30 min of annealing, and (d) cooling. Solid lines represent PP while the dash lines represent

PPOH...... 57

Figure 3.5 Real-time evolution of IR absorbance for the crystalline peak at 1045cm-1 and the amorphous peak at 2721 cm-1 during the 4 stages of processing cycle (heating + 1 min. holding, stretching, annealing and cooling) for (a) PP processed 110°C, and (b)

PPOH processed at 99°C...... 59

Figure 3.6 Orientation function of amorphous and crystalline segments in PP during the four stages of processing cycle: (a) heating + 1 minute of holding, (b) stretching, (c) 30 minutes of annealing, (d) and cooling. The stars represent the crystalline orientation functions of the final samples at room temperature, and calculated from offline WAXD measurements using equation (3)...... 60

Figure 3.7 Orientation function of the amorphous and crystalline segments in PPOH during the four stages of processing cycle: (a) heating + 1 minute of holding, (b) stretching, (c) 30 minutes of annealing, (d) and cooling. The stars represent the crystalline orientation functions of the final samples at room temperature, and calculated from offline WAXD measurements using equation (3)...... 60

Figure 3.8 DSC endotherms for (a) PP and (b) PPOH after heating to the deformation temperature, holding for a minute and cooling back to RT (heat–cool)...... 62

Figure 3.9 (a) Onset of melting, and (b) Difference between the primary and secondary melting temperatures as a measure of lamella-thickening, for PP (filled symbols, dashed lines) and PPOH (open symbols, dotted lines) polymers subjected to different deformation temperatures. The lines are drawn for clarity...... 62

Figure 3.10 (a) SAXS intensity profile as a function of the scattering wave vector. Solid curves correspond to heat-anneal-cool specimens of PP. Dotted curves correspond to specimens that were stretched until the point of minimum birefringence in Figure 3. 4(b) and cooled. The curves are shifted vertically for clarity. The inset is a 2D profile for the heat-anneal-cool specimen of PP at 110C indicating random orientation. (b) Long- spacing of lamellar crystals of PP (filled symbols, dashed lines) and PPOH (open symbols, dotted lines) subjected to different deformation temperatures. The lines are drawn for clarity.The long-spacing of underformed PP and PPOH is shown by the horizontal dashed lines...... 64

Figure 3.11 Stress-optical curves for PP and PPOH polymers at different deformation temperatures for the stretching stage. Different regimes for PP stretched at 110C are marked and separated by the dotted lines...... 71

Figure 3.12 Orientation functions for crystalline and amorphous polymer chain segments as a function of true stress for (a) PP and (b) PPOH at different deformation temperatures.

Different regimes corresponding to Figure 11 are marked and separated by the dotted lines for (a) PP stretched at 110C, and (b) PPOH stretched at 99C. The inset is a magnification of the selected area on the plot...... 71

xxi

Figure 3.13 Wide-angle X-ray diffractograms of PP and PPOH polymers after a full processing cycle (heating + 1 min holding, stretching, annealing and cooling). The inset shows a 2D WAXD pattern of PP stretched at 110C...... 77

Figure 3.14 Schematic depicting the structural evolution of polypropylene during uniaxial stretching from the partially molten state...... 78

Figure 4. 1 Synthesis of PI (BTDA-DAH) polymer. [119] ...... 84

Figure 4.2 DSC thermograms for PI (BTDA-DAH) polymer obtained from solution-cast and dried films. The dashed lines indicate the ideal stretching window in which the thermal crystallization rates are very low. The symbols indicate the following: Tg – glass transition temperature, Tcc – cold crystallization temperature, Tm – melting point., ΔHc

– heat of cold crystallization, and ΔHf – heat of fusion...... 88

Figure 4.3 True stress-Hencky strain plots for PI(BTDA-DAH) films at various temperatures. All the films were heated to stretching temperatures and isothermally maintained at that temperature for 5 minutes before stretching at a constant rate of

0.0024sec-1...... 89

Figure 4.4. True stress-Hencky strain plots for PI(BTDA-DAH) films at various rates. All the films were heated to 105°C and isothermally maintained at that temperature for 5 minutes before stretching...... 90

Figure 4.5 True stress-birefringence behavior of solution cast Polyimide(BTDA-DAH) films during uniaxial deformation at a constant rate of 0.0024sec-1 for a) varying temperatures and b) different regimes of stress birefringence behavior. All the films were

xxii

heated to stretching temperatures and isothermally maintained at that temperature for 5 minutes before stretching...... 93

Figure 4.6 True stress-birefringence behavior of solution cast PI(BTDA-DAH) during uniaxial deformation at a constant temperature of 105°C and varying rates. All the films were heated to 105°C and isothermally maintained at that temperature for 5 minutes before stretching...... 94

Figure 4.7 Critical values of true stress and birefringence at the point of deviation from linearity of stress optical rule of solution cast PI(BTDA-DAH) films during uniaxial deformation at a constant rate of 0.0024sec-1 and varying temperatures...... 94

Figure 4.8 Photoelastic and Regime I slope as a function of temperature of solution casting PI(BTDA-DAH) films during uniaxial deformation at stress rate 0.0024sec-1. ... 95

Figure 4.9 Hencky strain-birefringence behavior of solution cast PI(BTDA-DAH) films during uniaxial deformation at a constant rate of 0.0024sec-1 and for varying temperatures. All the films were heated to stretching temperature and isothermally maintained at that temperature for 5 minutes before stretching...... 97

Figure 4.10 Hencky strain-birefringence behavior of solution cast PI(BTDA-DAH) films during uniaxial deformation at a constant temperature of 105°C and for varying stretch rates. All the films were heated to stretching temperature and isothermally maintained at that temperature for 5 minutes before stretching...... 97

Figure 4.11 Mechano-optical behavior of solution cast PI(BTDA-DAH) films during uniaxial deformation at a constant temperature of 105°C, strain rate of 0.0024sec-1, and different stretch ratios with correspondent WAXD patterns and % of crystallinity...... 100

xxiii

Figure 4.12 DSC thermogram curves for PI(BTDA-DAH) polymer films. Samples stretched uniaxially at different stretch ratios, a constant temperature of 105°C, and a strain rate of 0.0024sec-1...... 102

Figure 4.13 DSC thermogram curves for PI(BTDA-DAH) polymer films. Samples stretched uniaxially at different stretch temperatures at a constant strain rate of 0.0024sec-

1...... 102

Figure 4.14 Schematic representation of the structural evolution of PI(BTDA-DAH) during uniaxial stretching...... 105

Figure 5.1 Synthesis of PI (BTDA-DAH) polymer...... 112

Figure 5.2 DSC thermogram curves for PI (BTDA-DAH) polymer obtained from undeformed film. The dash marks indicate the stretching window in which the thermal crystallization rates are slow. The symbols indicate the following: Tg – glass transition temperature, Tcc – cold crystallization temperature, Tm – melting point, ΔHc – heat of fusion of cold crystallization peak, and ΔHf – heat of fusion,-amount of crystallinity.

...... 117

Figure 5.3 Plot of ퟏ푻풎 − ퟏ푻풎풐흑ퟏ vs 흑ퟏ푻풎 for Polyimide (BTDA-DAH) with m- cresol...... 118

Figure 5. 4 True stress-Hencky strain plots for PI(BTDA-DAH) films during the stretching and relaxation process. Samples were stretched between 10% to 250% beyond their original length at 100°C and relaxed for 10 minutes. In the inset, engineering mechanical behavior is presented; the spontaneous deformation cannot be observed. .. 119

xxiv

Figure 5.5 Normalized true strain as a function of relaxation time for PI(BTDA-DAH) relaxed at 100°C for 10 minutes...... 121

Figure 5.6 Normalized birefringence as a function of relaxation time for PI(BTDA-DAH) relaxed at 100°C for 10 minutes. The values indicate the engineering strain at relaxation time zero...... 122

Figure 5.7 Real-time evolution of IR absorbance for the C=O asymmetric peak at

1770cm-1 during stretching and relaxation ...... 125

Figure 5.8 PI(BTDA-DAH) chain orient function and its dipole transition angle in a uniaxially oriented polymer...... 126

Figure 5.9 Normalized orientation function as a function of relaxation time for PI(BTDA-

DAH) relaxed at 100°C for 10 minutes. The values indicate the engineering strain at relaxation time zero...... 126

Figure 5.10 The mechano-optical behavior of PI(BTDA-DAH) during the stretching and relaxation process. Samples were stretched between 10% and 250% beyond their original length at 100°C and 10mm/min and relaxed for 10 minutes...... 129

Figure 5.11 The mechano-rheo optical behavior of PI(BTDA-DAH) during the stretching and relaxation process. Samples were stretched between 10% and 250% beyond their original length at 100°C and 10mm/min and relaxed for 10 minutes...... 131

Figure 5.12 Mechano-optical behavior of PI(BTDA-DAH) during the stretching and relaxation process with corresponding WAXD pattern. Samples are stretched at 100°C and 10mm/min and relaxed for 10 min. The values indicated in the pictures are referring to the level of deformation during uniaxial stretching and to the levels of crystallinity.

...... 134

xxv

Figure 5.13 DSC thermograms of stretched and relaxed PI(BTDA-HDA) samples which have been stretched at different deformation levels at 100°C and a rate of 10mm/min and then relaxed for 10 min...... 136

Figure 5.14 Percentage of crystallinity of PI(BTDA-DAH) as a function of true stress.

The circles represent stretched specimens and the triangles represent specimens that have been stretched and relaxed. all the samples are stretched at different strain deformations during uniaxial deformation at 100°C and at a rate of 10mm/min...... 137

Figure 5.15 Percentage of crystallinity of PI(BTDA-DAH) as a function of true (Hencky)

Strain. The circles represent stretched specimens and the triangles represent specimens that have been stretched and relaxed. all the samples are stretched at different strain deformations during uniaxial deformation at 100°C and at a rate of 10mm/min...... 138

Figure 5.16 Detailed mechano-optical behavior, orientation function behavior WAXD patterns, and dsc thermograms of PI(BTDA-HDA) stretched to 150% beyond their original length and relaxed for 1 h...... 141

Figure 5.17 Detailed mechano-optical behavior, orientation function behavior WAXD patterns, and DSC thermograms of PI(BTDA-DAH) stretched to 100% beyond their original length and relaxed for 1 h...... 142

Figure 5.18 Schematic representation the structural evolution of PI(BTDA-DAH) during relaxation after uniaxial stretching...... 144

Figure 5.19 The effect of relaxation time on the dielectric constant of PI(BTDA-DAH) relaxed specimen films which were stretched 150% at different frequencies. The test was made at 25°C...... 145

xxvi

Figure 5.20 The effect of relaxation time on the D-E loops of PI(BTDA-DAH) relaxed specimen films which were stretched 150% at different frequencies. The test was made at

25°C...... 146

Figure 6.1 Synthesis of PI (BTDA-DAH) polymer...... 151

Figure 6.3 Evolution of Hencky strain during sequential biaxial stretching at 90°C and different rates and stretching ratios...... 156

Figure 6. 4 change in the 24 dots during UCW...... 157

Figure 6.5 Evolution of Hencky strain during simultaneous biaxial stretching at 90°C and different rates and stretching ratios...... 157

Figure 6.6 Evolution of the in-plane and out-of-plane birefringence during simultaneous biaxial stretching at 90°C, different rates and stretching ratios...... 159

Figure 6.7 Evolution of the in-plane and out plane birefringence during sequential biaxial stretching at 90°C and different rates and stretching ratios...... 161

Figure 6.8 Refractive indices of PI(BTDA-DAH) films simultaneous biaxial stretched to a different stretching ratios at 90°C and 50 mm/min...... 162

Figure 6.9 Refractive indices of PI(BTDA-DAH) films sequential biaxial stretched to a different stretching ratios at 90°C and 50 mm/min...... 164

Figure 6.10 Strain optical behavior of simultaneous biaxial stretched PI(BTDA-DAH) films to a different stretching ratios and rates at 90°C...... 166

Figure 6.11 Strain optical behavior of sequential biaxial stretched PI(BTDA-DAH) films to a different stretching ratios and rates at 90°C...... 168

Figure 6.12 Schematic representation of the structural evolution of PI(BTDA-DAH) during simultaneous biaxial stretching...... 171

xxvii

Figure 6.13 Schematic representation of the structural evolution of PI(BTDA-DAH) during sequential biaxial stretching...... 172

Figure 6.14 The effect of frequency, temperature and stretching ratios on dielectric constant and loss of simultaneous biaxial stretched PI(BTDA-DAH) films...... 174

Figure 6.15 The effect of frequency, and stretching types and ratios on dielectric loss of biaxial stretched PI(BTDA-DAH) films which were compression vacuum molded after solution casting...... 175

Figure 6.16 The effect of frequency, and stretching types and ratios on dielectric constant of biaxial stretched PI(BTDA-DAH) films which were compression vacuum molded after solution casting...... 176

Figure 6.17 The effect of stretching ratios on dielectric breakdown of simultaneous biaxial stretched PI(BTDA-DAH) films...... 177

xxviii

KEYWORDS

Birefringence, capacitor, dichroism, mechano-optical behavior, orientation, polypropylene, polyimide, real-time URS-FTIR, rheo-optical properties, PPOH,

PI(BTDA-DAH)

xxix

LIST OF ABBREVIATIONS

PPOH- Hydroxyl-functionalized polypropylene

BOPP- Biaxially oriented polypropylene

PP- Polypropylene

MD- Machine direction

TD- Transverse direction

R.T- Room temperature

WAXD- Wide-angle X-ray diffraction

SAXS- Small-angle X-ray scattering

FTIR- Fourier transform infrared spectroscopy

IR- Infrared

PI(BTDA-DAH)- Polyimide(3,3′,4,4′-benzophenone tetracarboxylic dianhydride and 1,6- diaminohexane)

P(VDF-HFP)- Poly(vinylidene fluoride-co-hexafluoropropylene)

NMP- N-Methyl-2-pyrrolidone

PTFE- Polytetrafluoroethylene

DSC- Differential scanning calorimeter

PLA- Poly(lactic acid)

PET- Poly(ethylene terephthalate)

xxx

PPS- Poly(p-Phenylene Sulfide)

PEN- Poly(ethylene naphthalate).

EM- Electromagnetic

URS- Ultra-rapid-scan

PVA- Poly(vinyl alcohol)

PE- Polyethylene

SOR- Stress optical rule

UCW- Uniaxail Constant Width

SOR- Stress Optical Rule

xxxi

CHAPTER I

INTRODUCTION

One of the most demanding topics is finding an alternative energy source that can replace the needs of using oil. [1] The new source needs to generate energy that can be stored into energy storage systems. Batteries, fuel cells, flywheel, and capacitors are some examples of energy storage systems. The improvement of energy storage systems is a crucial part for future application such as: electric/hybrid vehicles, electromagnetic armor, electromagnetic rail gun, particle beam accelerators, and more. Energy storage systems are characterized by the amount of energy that they can store and the system's rate of energy transfer.

One of the energy storage systems that attracts researchers is capacitors; it is because of their relative short charge time and high ability to transfer energy. [2] Most of the conventional polymer capacitors are made from thin film. Therefore, the polymers are subjected to melt casting or solution casting process. If the thin film polymer is processed by a melt casting procedure, a tenter frame and film blowing machine are typically used while a roll-to-roll machine is used for solution casting process. During processing in these machines, the polymer films are subjected to a multi stage process, such as uniaxial stretching, biaxial stretching, heat setting, etc. The processing conditions play a critical role

in the final performance of the polymers. Due to lack of laboratory machinery, the mechanical, optical, and structural behavior at solid, rubbery, and partial molten state are still not completely understood.

The main goals in industry is to understand the effect of processing condition on the structural evolution of polymer in order to achieve specific polymer morphology. In order to address a solution for this subject, a real time measuring technique with fast measuring resolution in microseconds was used. Such machines are expensive and delicate; therefore, they are not accessible to most industry. A real-time spectral birefringence and ultra-rapid-scan polarized FTIR measurement instruments were built in The University of

Akron labs in order to track very fast, structural changes during deformation.

This work is part of collaborative project through which the following universities are attempting to design advanced polymeric capacitor films. The universities that participate in this project are the University of Connecticut, Rensselaer Polytechnic

Institute, Pennsylvania State University, Columbia University, and The University of

Akron.

This dissertation focuses on the effect of processing techniques on structural evolution of novel polymers, which have been found to be good candidates for capacitor applications. Chapter 3 focuses on real-time infrared-mechano-optical behavior and structural evolution of hydroxyl-functionalized PP during uniaxial deformation. Chapters

4-6 focus on real-time mechano-optical and infrared-mechano-optical characterization of

PI(BTDA-DAH) material during uniaxial deformation followed by relaxation and biaxial deformation.

CHAPTER II

LITERATURE REVIEW

2.1 Energy Storage Systems

The inability of 20th century power stations to operate at all times encouraged researchers to develop energy storage systems. These systems preserve and store electrical energy, enabling its use when it is needed. [3] Although advanced technology has allowed power stations to be more reliable, power outages still result from high energy demands.

Furthermore, climate change has encouraged governments all over the world to used more

“green” renewable energy sources. [4,5,6] Therefore, the number of energy storage systems continues to increase. Usually, the energy storage systems are characterized by the amount of energy that they can store and the system's rate of energy transfer (power output). Figure

2.1 shows a Ragone chart, which classifies the different types of energy storage systems according to their ability to store energy and their power output.

FIGURE 2.1 REGONE PLOT FOR DIFFERENT ENERGY STORAGE SYSTEMS. [7] (adapted with permission from “Electropaedia ” http://www.mpoweruk.com/alternatives.htm. Copyright © 2005, Woodbank Communications Ltd.).

2.2 Types of Energy Storage Systems

2.2.1 Battery

A battery is a storage device that stores electricity in the form of chemical energy.

A battery contains electrochemical cells (electrolytes), as well as positive and negative electrodes. When a battery is connected to external voltage, an electrochemical reaction takes place which charges the battery by storing electrons directly in the electrodes. When the battery is connected to an electronic circuit, an electrochemical reaction occurs which generates a flow of electrons into the circuit. Batteries are thought of as outstanding storage systems due to their high energy densities, high energy capabilities, cycling capabilities,

long life spans, and low initial cost. However, they cannot operate at high power levels for long periods of time, and rapid discharge may shorten their life time. [2,3,8,9]

2.2.2 Fuel Cell

A fuel cell is an energy storage system that converts and stores electrical energy by using an electrochemical reaction. The electrochemical reaction in fuel cells is an oxidation-reduction reaction involving oxygen. Many believe that fuel cells will replace batteries: fuel cells have similar work outputs, but higher storage capacities. However, fuel cells have limitations which must be eradicated before they can grow in popularity. In order for fuel cells to become the most popular energy storage device, their efficiency must be improved by preventing a dependency on the amount of power drawn from it, their high response time must be decreased, and their price per unit must be reduced. [2,8,9]

2.2.3 Superconducting magnetic energy storage (SMES)

SEMS is a storage device which stores electric currents from electrical energy. A magnetic field, which stores electrical energy, is generated when an electric current flows through a superconducting coil. The advantages of SEMS are that it provides highly efficient storage, it has a fast response time, a it has a fast cycle life, and that most of its stored energy may be transferred during discharge. However, SEMS also has a few

limitations, including a high price per unit and a strong magnetic field which can be related to environmental issues. [2,3,9]

2.2.4 Flywheel

Flywheel is an energy storage and generation device which works by an angular momentum of a spinning mass. Energy is generated and stored when the flywheel rotor is spinning and accelerates. The flywheel transfers the energy when the rotor slows down and acts like a generator. The amount of stored energy in flywheel depends on the length, mass, and acceleration of the rotor. The advantages of flywheel includes its long life cycles and high efficiency. [2,3]

2.2.5 Capacitors

A capacitor is a passive electronic element which stores electrical charge in the form of an electrostatic field. [10] Capacitors can be used for different purposes: they can stabilize fluctuating DC voltages, stabilize AC voltages, create timing circuits, etc. The purposes of the capacitors dictate the capacitor structure and the dielectric material with which it is built. The basic structure of capacitors is a sandwich (parallel plate) of conductor layers (electrodes), with a dielectric layer in the middle (Figure 2.2). When a current is applied, one conductor layer is positively charged while the other is negatively charged. In

the electric field, the dielectric polarized molecules orient to the conductor layer of opposite charge. [11]

In general, the capacitor’s “capacitance”, which is the capacitor’s ability to store charge, is defined by the ratio between the maximum amounts of charge, q, that the capacitor can have per unit voltage:

푞 퐶 = (2.1) 푉

For parallel plate capacitors, the capacitance of a capacitor can also defined as :

휀 휀퐴 퐶 = 표 (2.2) 푑 where 휀표, 휀 are the permittivity of vacuum and relative permittivity of the dielectric medium respectively. A and d are the conductor surface area and the distance between the two conductors respectively.

Capacitors are assessed by their relative permittivity, dissipation factor (tan δ), loss factor and dielectric breakdown. A dielectric constant is a material property that is defined as the ratio of the capacitance of the material to that of the vacuum for the same geometry.

The dielectric constant determines the maximum electrostatic energy which can be stored per unit volume when unit voltage is applied. Capacitors with higher dielectric constants have lower volumes than those with lower dielectric constants with the same capacitance value. Usually, the dielectric constant of polymer is between 2 and 6. Tan δ determines the ratio of its equivalent series resistance (ESR) to reactance at a specific frequency. ESR is the summation of all resistance in the capacitor (the dielectric layer, the electrode, and the terminal leads at specific frequencies). Loss factor is defined as the ratio of real power to apparent power. It is determined as the change in the dielectric through dissipative phenomena such as conduction, slow polarization currents, etc. Dielectric breakdown is defined as the voltage required for electric breakdown. [11]

At the end of the 19th century, Maxwell developed his theory on the relationship between electric and magnetic phenomena. [12] In his theory, Maxwell used Ampere’s circuit law, Faeaday’s induction and Gauss’s law for magnetic and electric fields to describe the propagation of electromagnetic waves. In this theory, Maxwell described the relationship between the dielectric constant of non-polar dielectric materials and refractive index as:

휀 = 푛2 (2.3)

2.3 Interaction of the dielectric constant and Light

2.3.1 Refractive index

Light beam is considered as an electromagnetic wave because it combines propagation of electric and magnetic vectors. These vectors are perpendicular to each other when the light travels through homogenous isotropic media as can be seen in Figure 2.3.

When the light travels through polarized media, the magnetic and electric vectors’ orientation becomes disordered. When this happens, the light is considered to be non- polarized light. It is only when the magnetic and electric vectors have order that the light considered as polarized. Therefore, by controlling the structure (polarized) of media which light is travels through, allows one to control the type of light polarization.

FIGURE 2.3 ELECTOMAGNETIC VAVE. [13] (adapted with permission from ref. “electromagnetic radiation” by m. abramowitz, m. j. parry-hill, and m. w. davidson, https://micro.magnet.fsu.edu/primer/java/scienceopticsu/electromagnetic/index.html).

When a parallel beam light travels through transparent medium, the beam intensity decreases because part of the incident ray will reflect back onto the medium of the incident ray and part is transmitted to the new transparent medium. If the angles of transmitted ray and the incident ray with the normal to the medium surface are unequal, this meana the transmitted ray was refracted. Figure 2.4. shows an example of when the transmitted ray is refracted. By using Snell’s law, the relative refractive index can be calculated:

퐬퐢퐧 휽푰 풏ퟏퟐ = (2. 4) 퐬퐢퐧 휽푹

Where 푛12 is the relative refractive index, 휃 is the angle between the ray of the normal to the medium plate, while the donation I and R are the incident and refract ray.

FIGURE 2.4 REFLECTION AND REFRACTION OF A RAY. [14] Adapted from ref. [14]

2.3.2 Birefringence

An optical anisotropy medium is defined as a system whose refractive indices depend on the direction of the system. When an electromagnetic wave travels through an anisotropy medium, its two plane polarized waves travel in different speeds and directions

with the refractive indices principle directions. One of the rays is reflected while the other continues to travel on the same path. The reflected ray is called an “extraordinary” wave while the second wave is called “ordinary” as can be seen in Figure 2.5. The direction of both extraordinary and ordinary rays coincide with the principle refractive indices. [15]

The extraordinary and ordinary rays travel in the optic axis direction they exhibit the same speed. As a result of the different velocities between the extraordinary and the ordinary, there is phase difference between the two rays which is calculate by:

2휋Γ 훿 = (2.5) 휆 where 훿, 휆, Γ are the phase differences, wavelength and retardation respectively. The birefringence can be calculate by:

Γ 푛 = (2.6) 12 푑 where 푑, 푛12 are sample thickness and birefringence respectively.

FIGURE 2.5 PROPAGATION OF LIGHT THROUGHOUT THE CALCITE CRYSTAL WHICH IS BIREFRINGENT MEDIUM. [15] (adapted with permission from “Microscopy” by B. Murphy, K. R. Spring, T. D. Fellers , and M. W. Davidson, http://nikon.magnet.fsu.edu/print/articles/polarized/birefringence-print.html ).

2.3.3 Measurement anisotropy optical properties

Polymer anisotropy(birefringence) can be measured by direct and indirect optical measurements. The indirect methods usually involve measurement of the material refractive indices, while the direct methods usually involve measurement of light retardation and intensity.

2.3.3.1 Indirect methods

In the indirect methods of measurement, the material refractive indices are calculated by using Snell’s law (equation 2.4). When the first medium has a higher refractive index than the second medium and the θR is 90° then equation 2.4 can be rewritten as:

푛1 sin 휃푐 = 푛2 (2.7)

where θc is the critical angle that θR is 90°. By knowing θc and n1 the refractive index of the unknown medium can be calculated. Refractometer instruments use this rule to measure the refractive index of the unknown medium. One of the down false of this instruments are that it measures the refractive indices of the surface and not the bulk polymer, which may be different. [16] Samuels showed that the percentage of crystallinity of films, the birefringence in the principle direction, and the refractive index distribution in the plane of the film can be measured by using a refractometer such as the Abbe refractometer. [16]

Cakmak et al. described the effect of biaxial protocols on the refractive indices and birefringence of PET films. [17]

2.3.3.2 Direct methods

In the direct methods, the material anisotropy is calculated by measuring the wavelength retardation of light or light intensity which passes and propagates in an anisotropy material. Usually, when measuring the material anisotropy by a wavelength retardation method, a compositor is required. A polarized light at 45° to the one of the material principle axes, passes through the sample. As a result, the light is split into ordinary and extraordinary rays which are parallel to each other. The velocity of these rays can be calculated by:

푐 푣 = (2.8) 푛푖 were v is the light velocity, c is the speed of light in a vacuum, and n is the refractive index.

The notation i represents the plane direction of the sample. If the sample is isotropic, the refractive indices of samples in the x and y directions are the same. As a result, the velocities of the ordinary and extraordinary rays are equal. However, when the sample is anisotropy, the refractive indices of samples in x and y directions are different. As a result, there are phase differences between the two rays which is calculated by equation 2.5.

A compositor, which is at 90° with respect to the polarizer, is located beyond the sample. By adjusting the compositor direction in respect to the polarizer the phase

shift can be eliminated. The amount of the rotation in the compositor resolve the values of the retardation. Therefore, the birefringence can be calculated by equation 2.6. In the case of a sample of biaxial optical symmetry, the out of plane birefringence (푛23) can be calculated by Stein’s equation:

sin2 ∅ 훤 −훤 √(1− ) 1 0 ∅ 푛̅2 ∆푛23 = − ( sin2 ∅ ) (2.8) 푑0 푛̅2

where 푛̅ is average refractive index. ∅ is the tilted angle which the 훤∅ is measured.

When using the intensity method, the birefringence is calculated as a function of retardation of light which passes through the sample, as well as wavelength. Several techniques, such as single wavelength, dual wavelength, multi-wavelength, and on line spectral birefringence were developed which complemented the intensity method. [18-22]

Typically, the intensity method entails light passing through a polarizer, the sample, an analyzer, and a detector.

The effect of processing conditions and molecular parameters on stress optical behavior was studied in the past for polymer melts and concentrated solutions. It was found that the SOR is valid only in some cases. For example, Figure 2.6 shows the different cases when the SOR deviates from linearity. The first appears when a polymer film is stretched at a temperature close to its glass transition and a glassy state results. The second occurs when a polymer film is stretched to high extension and the polymer chain is high oriented; the stress starts to increase faster than the birefringence. The third takes place as strain induced crystallization occurs during stretching. [23]

FIGURE 2.6 MECHANO-OPTICAL BEHAVIOR OF DIFFERENT TYPE OF POLYMER DURING UNIAXIAL DEFORMATION. A DEVIATION FROM SOR DUE TO STRETCHING TEMPERATURE CLOSE TO TG. B. DEVIATION FROM S.O.R DUE TO HIGH ELONGATION CHAIN. C. DIVINATION FROM S.O.R DUE TO CRYSTALLIZATION. [23] (adapted with permission from “A birefringence study of polymer crystallization in the process of elongation of films” by D. S. Ryu, T. Inoue, K. Osaki. Polymer 1998, 12, 2515-2520. Copyright © 1998, Published by Elsevier Ltd.).

Figure 2.7 shows the relation between stress and birefringence for PS at temperatures between 120°C to 157°C and elongation rates from 0.075sec-1 to 2sec-1 as was found by Matsumoto et al. In this study it was found that the stress optical coefficient was a constant for all stretching temperatures. This was true until a critical stress when, beyond this point, the stress increased faster than the birefringence. [25]

FIGURE 2.7 MECHANO-OPTICAL BEHAVIOR OF PS WHICH WERE STRETCHED AT DIFFERENT STRETCHING TEMPERATURE. [24] (adapted with permission from “Stress Birefringence in Amorphous Polymers under Nonisothermal Condition” by T. Matsumoto, and D. C. Bogue, Journal of Polymer Science: Polymer Physics Edition Polymer 1977, 15, 1663-1674. Copyright © 1977, John Wiley & Sons, Inc.).

Ryu et al. studied the effect of stretching temperatures and rates on the stress optical rule of amorphous PET film during uniaxial deformation. They found that at low stretching temperature, or high stretching rates, the stress optical rule is invalid. They related this behavior to quickly increasing stress during the deformation. At higher stretching temperature and slower starching rate, the stress optical rule was valid until reaching a critical stretching ratio at which birefringence increases rapidly compared to stress. They found this deviation is a result of strain induced crystallization. When the films continue to be stretched at the same temperature and rates the stress suddenly increases faster than birefringence due to the non-gaussian nature of the highly extended chains. [23]

FIGURE 2.8 STRESS OPTICAL BEHAVIOR OF AMORPHOUS PET DURING UNIAXIAL STRETCHING. [23] (adapted with permission from “A birefringence study of polymer crystallization in the process of elongation of films” by D. S. Ryu, T. Inoue, K. Osaki. Polymer 1998, 12, 2515-2520. Copyright © 1998, Published by Elsevier Ltd.).

2.4 Polymer crystallization

Polymers are usually characterized by their ability to crystalize. Non crystallized polymers are considered amorphous polymers, while polymers that crystallize are called semi-crystalline polymers. The crystallization can take place by thermal crystallization, which usually occurs during cooling a polymer melted between the polymer’s glass at melt transition temperatures, or during heating a semi-crystalline polymer which is in an amorphous state between cold crystallization and melting temperatures. In some cases, crystallization can take place also during deformation by strain induced crystallization phenomena. Typically, thermal crystallization polymers are characterized by their crystallization rate: fast crystallization polymers and slow crystallization polymers.

2.4.1 Fast crystallization polymer

Polymers which belong to this crystallization group usually are steroregular polymers [25] which have flexible back bond chains that have a high level of architectural symmetry. Polymers in this group exhibit glass transition below room temperature.

Therefore, only under extreme rates of cooling can these polymers appear in their glassy state. Usually, these polymers are processed in the melt or partial melt state.

2.4.2 Slow crystallization polymer

Polymers which belong to this crystallization group usually have stiff backbone chains that can organize in a steroregular polymer. Polymers in this group exhibit three transition temperatures--glass, cold crystallization, and melting. [26] The glass transition of these polymers takes place above room temperature. In those cases, where the polymers are quenched bellow their Tg, their thermal crystallization is suppressed and the polymer appears in its amorphous phase. Usually slow-crystallizing polymers are processed in their rubbery state between the glass transition and cold crystallization transition, where they exhibit very low thermal crystallization rates due to high rubbery viscosity that suppress polymer chain diffusion. [27-29]

2.4.3 Thermal crystallization

Thermal crystallization occurs only when there is nucleation which will start the crystallization process, followed by growth in the crystal size by merging of chain segments from the amorphous phase. The nucleation can be heterogeneous which takes place when a foreign particle introduces itself into the matrix and starts the nucleation or homogenous phase in which spontaneous nucleation occurs. In order for nucleation to take place, the nucleus droplet size must to be greater than critical nucleus size. Under these conditions, the addition of polymer chains from the amorphous regime will decrease the

Gibbs free energy and the growth of nuclei is spontaneous. The rate of formation of nuclei is dependent on the rate of arrival of new molecules or polymer chains at the nuclei droplet surface. At low temperatures, the rate of nucleation will be zero due to fact that the polymer chain in the amorphous region cannot travel to the nuclei surface. At low temperatures, but higher than those of polymer glass transitions, the rate of nucleation will increase and the nuclei thickness will be thicker when compared to the ones which occur at higher cooling temperature.

Similar to nucleation, the growth velocity of the crystal is dependent on the cooling temperature. At low cooling temperature (close to Tg) the growth rate is very slow, and could be considered negligible. At high temperature Tm and above the polymer is melted.

The maximum rate of the crystallization growth is usually in between glass transition and melting point. [30,31] Magill studied the effect of cooling temperature and molecular weight on the crystallization of Poly(tetramethyl-p-silphenylene)siloxane spherulite as can be seen in Figure 2.9 In his study he found that the temperature of the maximum

crystallization rate is not dependent on the polymer molecular weight. However, he found that the molecular weight has significant effect on the crystallization rate. In his study,

Magill showed that the crystallization rate decreases with increasing the polymer molecular weight. The reason for this phenomenon is the amount of the entanglement that takes place in the polymer matrix. When the entanglements are high, which occurs in high molecular weight polymers, the polymer chain has more difficulty accommodating the necessary conformational rearrangement for growing the crystal length. [32]

FIGURE 2.9 GROWTH RATE OF POLY(TETRAMETHYL-P-SILPHENYLENE)SILOXANE SPHERULITE DURING CRYSTALLIZATION AS FUNCTION OF TEMPERATURE AND MOLECULAR WEIGHT. [32] (adapted with permission from “Crystallization of Poly‐ (Tetramethyl‐p‐Silphenylene)‐Siloxane Polymers” by J. H. Magill . Polymers. Journal of Applied Physics, 1964, 35, 3249-3259. Copyright © 1964 The American Institute of Physics).

2.4.4 Strain Induced Crystallization

Strain-Induced crystallization is typically associated with polymer processing because it can occur only when the polymer deforms. For example, it appears when stretching amorphous polymer in its glassy state, stretching semi-crystalline and cross-link polymers in their rubbery state, and when shearing linear polymer melts. [33] The crystalline structure which forms during strain-induced crystallization usually differs from those which form during thermal crystallization. Furthermore, the rate of crystallization during strain induced crystallization is much faster than during thermal crystallization. For example, the crystallization time which PET is crystallized under isothermal condition at

90°C is more than a day, while during strain-induced crystallization it takes a few minutes.

[34] Moreover, the activation energy needed for strain-induced crystallization to occur is lower than those for thermal crystallization. [33] Bourvellec et al. studied the kinetics of strain-induced crystallization of amorphous PET during uniaxial stretching. In their study, they assess the effect of strain rate and temperature on the rate of strain-induced crystallization. They found that strain rate controls the orientation of the amorphous region which controls the strain-induced crystallization rate. That is, at higher strain rate, the polymer chain crystallization increases, which in turn increases the rate of strain-induced crystallization. Furthermore, at high stretching temperature relaxation takes place which delays the beginning of crystallization, but does not effect the rate of the strain-induced crystallization. [34]

2.5 Polypropylene

PP is a semi-crystalline polymer which belong to fast crystallization polymer group.

[31] In fact it is one of the most commodity polymer in the world. [35] Due to its high mechanical, optical, physical, dielectric, and processability, as well as its low cost, PP can be found in a variety of applications, such as in packaging or capacitors, as well as automotive, medical, and consumer goods, etc. [36]

PP is polymerized by free radical reaction. By using a specific catalyst, PP tacticity can be controlled. That is, by using VCl4/Al(C2H5)Cl as catalyst, a syndiotactic PP will be polymerized, while when using TiCl3/ Al(C2H5)2Cl as catalyst, the PP will have been in isotactic tacticity. Typically, PP has a glass transition around 0°C and a melting point of around 160°C. PP can crystalize in three different crystal structures: α-form (monoclinic),

β- form (hexagonal), γ- form (orthorhombic). [37,38] α-form structure of isotactic PP is obtained under normal processing conditions. The α-form structure of isotactic PP is obtained under normal processing conditions. The α-form structure is reported by Natta and Corradino. [38] In this case, the polymer chains organize in helical structure in a monoclinic unit cell with the following dimensions: 20.8Ȧ , 6.6Ȧ , and 6.5Ȧ . per unit cell.

[38] Typically, the folded lamella thickness is between 50Ȧ to 200Ȧ . The lamella can grow in across-hatched structure which results in forming spherulites structures. [38] The β-form structure of isotactic PP is obtained under low isothermal crystallization temperature or by adding a specific β-form nucleation agent such as pimelic acid and calcium stearate. The polymer chain in the β-form organize in a unit cell of hexagonal structure. In β-form

structure the lamella form a sheaf-like spherulitic structure. The β-form can convert to α- form upon heating. [38] γ -form structure can be obtained in low molecular weight; isotactic PP under high pressure. The polymer chain in the γ -form organize in the orthorhombic structure of unit cell. [38]

PP is typically processed in melting state or in partial melting state. In melting state the polymer is 100% melted, while in partial melting state the polymer is only partially molten. The first is used in casting unorinted film process and fiber spinning, while the second is used when casting oriented film processed by a tenter frame machine and film flowing from unorinted film.

2.6 Polyimides

Polyimides are a class of high performance polymers whose high performance is due to their high heat resistance as well as their chemical, mechanical, electrical and physical properties. [39-41] As a result, PI can be found in variety of industries, such as aerospace, electronics, and optics, for applications, such as membranes. [42-45]

Polyimides can have aromatic, aromatic-aliphatic, aliphatic forms. [46,47] However, as of today, only the aromatic PI is commercially available. Polyimides can be synthesized by two methods: a two-step process via PAA and a one-step process. Figure 2.10 shows the two-step process. This first method is more common. In short, in the two-step method PAA is synthesis from dianhydride, diamine and dipolar solvent. In the first step of this process, the amino group in the diamine, nucleophilic attacks the carbonyl group in the form of dianhydride. In the second step, the PAA is imidized by thermal or chemical reaction.

[48,49] Typically, when thermal imidization is used, the final structure of the PI (the film, fiber, and coating) are made in the PAA form; and only then is the thermal reaction processed. [48] The one-step PI synthesis process can take place only when the PI are made soluble in organic solvent at polymerization temperature. The monomers are mixed with the solvent at high temperature which makes the imidization reaction proceed rapidly [48]

The reactivity of the monomers, the solvent’s acid-base interaction, and reaction conditions play an important role in the rate of the reaction, thermal stability, and amount of trapped solvent in the final product (its molecular weight, physical and mechanical properties, etc.).

[48-50]

FIGURE 2.10 SYNTESIS OF PI. [50] (adapted with permission from “Synthesis, Characterizations, Aging and Semiconductor Device Passivation” by S. Diaham, M. L. Locatelli, R. Khazaka, InTech Polymers. under CC BY 3.0 license. Available from: http://dx.doi.org/10.5772/53994).

2.7 Processing techniques for thin films in capacitor technology.

The processing techniques which are used for manufacture of polymer thin film for capacitor application are melt casting film, melt casting film followed by tenter frame machine, blown film, and solution casting. The final film properties are dependent on the processing techniques and conditions. The processing strategy to manufacture thin films is different for fast and slow crystallization polymers even when the processing technique is the same.

2.7.1 Melt Cast Film By Extrusion

The melt cast process is one of the methods most commonly used to manufacture polymer film. This method was developed in the early 20th century. The melt casting line is comprised of three major parts: extruder, die, and series of chill rolls. Using this technique, the polymer, which is typically in a solid state, is converted to melt inside the extruder and made to flow through a flat die in order to create a film structure. [51] The die lips grab and the speed of rotation and temperature of the chill rolls control the final film. At high rotation speed the film is stretched to MD which increases the film orientation. Therefore, the mechanical properties of the film are different in the MD and

TD. As a result of the uniaxial stretching the film thickness decreases to as little as ~7 µm.

2.7.2 Tenter-frame

Tenter-frame biaxial stretching is a processing method which has been industrially employed for stretching film in two directions. Although, this method has been industrially implemented for more than a decade, the behavior of semi-crystalline polymer chain segments in the amorphous and crystalline domains during such processing is still not well understood. Industries engaged the tenter-frame machine for thin film applications because of the superior mechanical properties of the final film in both MD and TD. In some cases, this process may improve gas permeability, dielectric properties, etc. That is, breakdown strength of BOPP from this method is higher when compared to melt casting. [52,53]

The tenter frame machine is comprised of five major parts, extruder, die, a series of godets rolls, a tenter clamps section, and take off film rolls. For the case in which the fast crystallization polymer is used, i.e. PP, the polymer is heated to a partially molten state before stretching, as can be seen in Figure 2.12; then stretching in the machine direction

(MD) is followed by annealing and cooling; afterward, reheating to a partially molten state is followed by stretching in the transverse direction (TD). Eventually the film is annealed, followed by cooling, and fed to the take-off stand. [54,55]

FIGURE 2.11 TEMPERATURE PROFILE OFFAST CRYSTALLIZATION POLYMER DURING BI- AXIAL PROCESSING CYCLING BY TENTER FRAME MACHINE. (A) TENTER FRAME MACHINE, (B) TEMPERATURE PROFILE OF PP, AND (C) THERMAL BEHAVIOR (DSC) OF FAST CRYSTALLIZATION POLYMER.

For the case in which a slow crystallization polymer is used, i.e. PET, the polymer melt is quenched after it flows through the die below the polymer glass transition in order to prevent crystallinity in the film. The polymer film is heated to a temperature between glass transition and cold crystallization temperature before stretching, as can be seen in

Figure 2.12; then stretching in the machine direction (MD) and cooling; afterward reheating to a temperature between glass transition and cold crystallization temperature is followed by stretching in the transverse direction (TD). Eventually the film is annealed, followed by cooling, and fed to the take-off stand. [54,55]

Stretching window

FIGURE 2.12 TEMPERATURE PROFILE OF SLOW CRYSTALLIZATION POLYMER DURING BI- AXIAL PROCESSING CYCLING BY TENTER FRAME MACHINE. (A) TENTER FRAME MACHINE, (B) TEMPERATURE PROFILE OF PP, AND (C) THERMAL BEHAVIOR (DSC) OF SLOW CRYSTALLIZATION POLYMER.

2.7.3 Blown Film

Blown film is a processing method which has been industrially employed for stretching film in two directions. The blown film process is comprised of five major parts, extruder, annular die, air ring, series of pinch and guide rolls, and take off film rolls, as can be seen in Figure 2.13. In this process the polymer melt flows through the annular die. Then the film is extensionally stretched by the series of guide rolls over a mandrel of air trapped inside the blown film bubble, while the polymer is cooled by external air rings, as well as internal bubble cooling distributors in many cases. By controlling the process conditions, it is possible to produce balance orientation as well as unbalance orientation.

Xu et al. studied the effect of film processing methods on the dielectric breakdown of PP. In their study they found that the BOPP which was made by the blown film method had lower breakdown strength. Furthermore, they found that the final film thickness that can be achieved using the tenter-frame machine was lower than through film blowing, 3

μm as compared to 10 μm. [53]

FIGURE 2.13 SCHEMATIC OF BLOWN FILM PROCESS. [56] (adapted with permission from “Production of films, containers, and membranes“ by W. McKeen in Permeability Properties of Plastics and Elastomers: A Guide to Packaging and Barrier Materials”. Copyright © 2012, Elsevier Inc.).

2.7.4 Double Blown Film

The blown film method is more suitable for fast crystallization polymers. Therefore, a modification in the blown film process was made to adjust the blown film method for slow crystallization polymers. The double blown film line is similar to blown film with the

addition of another annular die, and air ring. Using this method, the polymer melt is quenched after it flows through the die below the polymer glass transition in order to prevent crystallinity in the film. The bubble film is heated to a temperature between glass transition and cold crystallization temperature before stretching, as can be seen in Figure

2.14. Eventually the film is annealed, followed by cooling, and fed to the take-off stand.

[57]

FIGURE 2.14 SCHEMATIC OF DOUBLE BLOWN FILM PROCESS. [57] (Reprinted with permission from “Films” by T. Brink, http://extrusionist.com/extrusion- processes/films/index.html ).

2.7.5 Solution Casting

Solution casting was one of the first methods for manufacturing film. After the development of the extruder, the usage of this method was dramatically reduced. With the solution cast method, the final film quality and the film thickness homogeneity is high.

Therefore, this process became popular again. For this process, solution is prepared by

mixing polymer with a good solvent. Then the solution is cast on a substrate which moves by belt. Finally, the solidification phenomenon starts due to evaporation of the solvent.

2.7.6 Processing of fast crystallization polymers in partial molten state.

For fast crystalline polymers, i.e. PP, film is typically melt cast. It is first cast into an unoriented crystalline state and then it is reheated to a temperature at which the film is partially molten. [58] The effect of uniaxial and biaxial stretching of PP in a partial molten state were intensively studied in the past. Koike et al. study the structural development of

PP during stretching and holding at the partial molten state by using the real time spectral birefringence technique. [58] In their study they revealed that the stress optical rule is valid in the partial molten state. In the partial molten state, the birefringence almost stays constant during stretching in the elastic region. After the yielding point, the birefringence started almost linearly increase with stress until the point at which strain hardening started.

Furthermore, they found the stress level at which the birefringence starts to increase depends on the stretching temperature and rates in the partial molten state as can be seen in Figure 2.15. At higher stretching temperatures, the birefringence starts to increase at lower stress values. Also at low stretching rates the birefringence starts to increase at lower stress levels. [58]

FIGURE 2.15 MECHANO-OPTICAL BEHAVIOR OF PP DURING STRETCHING AT ITS PARTIAL MOLTEN STATE. [58] (adapted with permission from “Real time development of structure in partially molten state stretching of PP as detected by spectral birefringence technique” by Y. Koike, M. Cakmak. Polymer 2003, 44, 4249-4260. Copyright © 2003, Elsevier Science Ltd.).

Samuels studied the effect of uniaxial deformation on the morphology of PP films at the partial molten state. He found that during the early stage of uniaxial stretching the birefringence decrease, and at higher stretching levels the birefringence increases one again. He explained this behavior as result of lamellar slip leading to c-axis orientation from a-axis orientation and separation of the lamella. During these two behaviors the polymer crystal first orients to the TD and then it orients to MD. [59]

Okajima et al. studied the effect of different type of biaxial stretching methods on morphology, chain orientation, and optical properties of PP in its partial molten state. [60-

67] In one of their studies, they investigate the effect of uniaxial stretching to TD while in a partial molten state after the film was uniaxial stretching in MD, and cool down on the chain orientation by using indirect method (abbe refractometer). During the uniaxial

stretching in MD the polymer chain oriented to the MD. Then during stretching to the TD the polymer chains reorient to the TD direction. In this experiment Okajima et al. found that the second stretching almost didn’t affect the polymer orientation in the normal direction. [60] In a different study Okajima et al. found that during polyaxially stretching there was a leg in the chain orientation of the crystal and amorphous regions of PP specimen as can be seen in Figure 2.16. [61] Furthermore, at higher stretching temperature the orientation decreased.

FIGURE 2.16 HERMAN ORIENTATION FUNCTION OF CRYSTALLINE AND AMORPHOUS REGIONS DURING POLYAXIALLY STRETCHING. THE ORIENTATION FUNCTION WAS MEASURED IN RESPECT TO THE NORMAL DIRECTION. [61] (adapted with permission from “Studies on the Stretching of Polypropylene Film. II. Polyaxial Stretching” by S. Okajima, S. Homma. Journal of Applied Polymer Science 1968, 12, 411-423. Copyright © 1968, John Wiley & Sons, Inc.).

The effect of simultaneous and polyaxially biaxial stretching on the morphology of a PP specimen was studied. [62] It was found that during the early stage of stretching,

the spherulites started to break into fibril structures regardless of the type of biaxial method. [62]

2.7.7 Processing of slow crystallization polymers between glass transition and cold crystallization temperatures.

When slow crystalline polymers (i.e PLA, PEN and PEEK) are shaped into a thin film, they are quenched after they flow through the extruder die. Therefore, their thermal crystallization is suppressed and the polymer appears in its amorphous phase. Usually slow-crystallizing polymers are processed (deformed) in their rubbery state between the glass transition and cold crystallization transition. They exhibit very low thermal crystallization rates in this state because their high rubbery viscosity suppresses polymer chain diffusion. [27-29] During stretching in the rubbery state, the polymer chains orient in the stretching direction.

Mulligan et al. studied the mechano-optical behavior of melt-cast amorphous PLA film and its structural evaluation during uniaxial deformation in the rubbery state. [27] In this study, Mulligan et al. revealed that when PLA films were stretched at stretching temperatures above Tll and slow stretching rates, three regimes of stress-optical behavior were observed. During regime I, which occurred in the early stage of deformation, the stress optical rule was observed; birefringence linearly increased with stress. During regime II the birefringence rapidly increased while the stress slowly increased. During regime IIIc stress rapidly increased while the birefringence reached a plateau. When PLA films were stretched at stretching temperatures below Tll and/or slow stretching rates, two regimes of stressoptical behavior were observed. During regime I, which occurred in the

early stage of deformation, the stress optical rule was observed. During regime IIIa, stress rapidly increased while the birefringence reached a plateau. [27] In this study, Mulligan et al. showed that the mechano-optical behavior depends on the ability of the polymer chains to relax during stretching. Therefore, at high stretching rates, the polymer chains do not have enough time torelax and mechano-optical behavior follows the I-IIIa regime scheme, as can be seen in Figure 2.17

FIGURE 2.17 TRUE STRESS-BIREFRINGENCE PLOTS FOR PLA FILMS STRETCHED AT (A) 60%/MIN AT VARIOUS TEMPERATURES. (B) 240%/MIN AT VARIOUS TEMPERATURES. (C) AT 960%/MIN AT VARIOUS TEMPERATURES. [27] (adapted with permission from “Nonlinear Mechanooptical Behavior of Uniaxially Stretched Poly(lactic acid: Dynamic Phase Behavior” by J. Mulligan, M. Cakmak. Macromolecules 2005, 38, 2333-2344. Copyright © 2005, American Chemical Society).

Martins et al. studied the effect of stretching temperature in the rubbery state, stretching rate, and molecular weight on the mechano-optical behavior of PEN films. [28]

The results which were observed in this study were similar to those found by Mulligan et al. Martins et al. found that the slop of regime I was independent of molecular weight, rate, and temperature with the exception of temperatures close to the glass transition temperature. Furthermore, a fast increase in birefringence and slow increase in stress in regime II was a result of a spontaneous deformation due to a coil to-stretch transition that also accompanies the formation of strain-induced crystallites. Finally, the increase in the stress with almost no change in birefringence in regime III is a result of the polymer chains having reached their limit of extensibility. [28] A structural model representative of PENs behavior during rubbery state deformation can be seen in Figure 2.18.

FIGURE 2.18 STRUCTURAL MODEL REPRESENTATIVE OF PENS BEHAVIOR UNDER RUBBERY STATE DEFORMATION. [28] (adapted with permission from “Large Deformation Mechano-Optical and Dynamical Phase Behavior in Uniaxially Stretched Poly(ethylene naphthalate)” by C. I. Martins, M. Cakmak. Macromolecules 2005, 38, 4260-4273. Copyright © 2005, American Chemical Society). 36

2.8. FTIR Spectroscopy

FTIR is a vibrational spectroscopy characterization technique which determines the morphology and microstructure of materials. This method has also been used for identification of materials’ compositions. The vibrational spectra of polymers are sensitive to changes in environmental effects near the groups that are of interest. As a result, FTIR is a good tool for study of molecular orientation, molecular conformation of multicomponent polymers such as block copolymer, polymer blend, polymer and filler, and semi- crystallization polymer.

The FTIR interferometer is the most important part in the FTIR instrument. The most basic and common FTIR interferometer is Michelson FTIR interferometer. However, all the FTIR instruments work the same way. A light beam is split by a beam splitter into two different beams which travel in two different directions. The beams then recombine at the beam splitter after they are reflected back from two mirrors, one of which is stationary while the other is a moving mirror. Based on the different path length of the beams, the interferogram, which is the combined beam, can be destructive or constructive. The interferogram then continues to travel through a sample and afterward travels to the detector. Figure 2.19 illustrated the Michelson FTIR interferometer mechanism.

FIGURE 2.19 MICHELSON FTIR INTERFEROMETER. [68] (adapted with permission from “Fourier Transform Infrared Spectrometry” by P. R. Griffiths, J.A.De Hasetch, J. D. Winefordner. Copyright © 2007, John Wiley and Sons).

Polarized IR dichroism is a useful technique which allows detection and study of the polymer anisotropy by investigating the Herman’s orientation function of multicomponent polymers. Splitting the IR beam into two components, one parallel and another perpendicular to the deformation direction, allows for calculation of Herman’s orientation function (fIR) using the following equation, [69]

(퐷−1)(퐷0+2) 푓퐼푅 = (2.9) (퐷+2)(퐷0−1) where D is the dichroic ratio, defined as the ratio of IR absorbance in parallel and perpendicular directions (A∥/ A⊥), and D0 is the dichroic ratio of an ideally oriented polymer, defined as

2 퐷0 = 2 cot 휓 (2.10)

where ψ is the transition moment angle of a specific side-group vibration with respect to the polymer chain axis.

In order to investigate a fast complex behavior of different components in a multicomponent polymer matrix system, real time polarized rheo-optical instruments were built. A real-time mechanical and FTIR spectroscopic measurement instrument was developed by Yeh et al. (70) This instrument allows one to track the changes of polymer systems during uniaxial stretching followed by relaxation. This machine could not measure parallel and perpendicular polarizability simultaneously. The parallel and perpendicular polarized IR beams were measured with a time lag of 50 ms from each other. Shigematsu et al. developed a real-time mechanical polarized IR spectroscopic measurement instrument which can measure stress strain and IR dichroism. [71] Unlike the Yeh et al instrument, this machine could measure the IR dichroism without any time lag. This instrument generated one spectrum in 3 seconds at 4cm-1 resolution. These instruments suffered from slow data acquisition and low resolution. Therefore, it was almost impossible to detect fast changes in the millisecond range. By changing the linear mechanism that gets

ZPD in the Michelson IR interferometer to a tilted/wedged circular spinning mirror,

Manning et al. designed an IR interferometer which could generate data acquisition rates of 1 millisecond at 4 cm-1 spectral resolution. [72] Based on the Manning et al. IR interferometer design, a URS-FTIR interferometer coupled with uniaxial stretcher and birefringence platform was built in our lab in 2010.

The URS-FTIR instrument is a sequence of components which allow the increase of resolution (scan speeds) of a Michelson FTIR interferometer. Details of the URS-FTIR interferometer are discussed elsewhere. [68] In short, the URS-FTIR had 8 major

components: an HeNe laser, IR source, beam splitter and compositor, cube corner mirrors, terminal mirrors, optical lenses, disk mirrors, and detector. In this system two lights beams were excited: one from a visible red light as a reference beam (HeNe laser) and an IR beam.

The IR beam's travel path is shown in Figure 2.20 By using different optical lenses, both beams were made to travel in parallel to each other to the beam splitter. The beam splitter split the beams into two paths, 50 % of the beams' intensity was reflected to the first disk mirror while the other 50% was transmitted to the second disk mirror. Both of the beams that were refracted and transmitted from the beam splitter had traveled similar paths. Both beams traveled to the rotational mirror. The beams' shape changed from spot-like to cone- like due to the variation in thickness of the disk mirror (beams swept out in an arc creating a circle on the surfaces). Then the beams traveled to the cube corner which translated and reflected the beams back to the disk mirror. From the disk mirror the beams traveled to the terminal mirror, which reflected the beams exactly back on the same path they had come.

Therefore, the beams traveled back to the disk mirror followed by the cube corner and then reflected back to the disk mirror which eventually traced the beams to the beam splitter.

Both of the beams, the refracted and the transmitted, recombined at the beam splitter and continued to traveled to the terminal stationery mirror which reflected and traveled to the sensor and to the detector.

FIGURE 2.20 TOP VIEW OF URS-FTIR WITH THE RED LASER BEAM TRAVEL PATH.

The machine had the capabilities to generating simultaneous parallel and perpendicular IR spectra up to 1000 scan/second. By using this instrument fast physical and chemical changes in multi-component polymer systems, such as semi-crystalline polymers, block copolymers, polymer blends, materials containing residual solvents, etc., could be studied. To give a few examples, the dynamics and molecular mechanisms of the complex thermal imidization process, the structural evolution of crystalline and amorphous parts of PP and PPOH during heating, uniaxial stretching, annealing and cooling, the behavior of the hard and soft segments of polyurethane(urea) during stretching and relaxation, and others were revealed by this machine. [54,73,74]

The polarized FTIR dichroism method of characterization is an important technique to understand the different behavior of crystalline and amorphous regions in semi- crystalline polymers during processing. Figure 2.21 shows Samuel's results about the effect of uniaxial deformation on crystalline and amorphous segment of PP film. In his study he found that the orientation function of the amorphous segment increases slowly with the deformation, while the crystalline regime increases much faster. [75] Ruiying et al. studied the effect of stretching temperature on crystalline orientation in PP. The result can be seen

in Figure 2.22. In this study they observed that the orientation function of the crystallinity region increases with increase in stretching temperature. [76]

FIGURE 2.21 ORIENTATION FUNCTION OF PP DURING UNIAXIAL STRETCHING AT 29°C. 풇풂 - ORIENTATION FUNCTION OF AMORPHOUS REGIME, 풇풄 - ORIENTATION FUNCTION OF CRYSTALLINE REGIME, AND 풇풂풗- AVERAGE ORIENTATION FUNCTION OF CRYSTALLINE AND AMORPHOUS REGIMES. [75] (adapted with permission from “Infrared dichroism, molecular structure, and deformation mechanisms of isotactic polypropylene” by R. J. Samuels. Die Makromolekulare Chemie Supplement 1981, 3, 241- 270. Copyright © 1981, John Wiley & Sons, Inc.).

FIGURE 2.22 ORIENTATION FUNCTION OF CRYSTALLINE REGION OF PP DURING UNIAXIAL STRETCHING DIFFERENT STRETCHING TEMPERATURE. [76] (adapted with permission from “Deformation-induced morphology evolution during uniaxial stretching of isotactic polypropylene: effect of temperature” by R. Bao, Z. Ding, G. Zhong, W. Yang, B. Xie, M. Yang, Colloid and Polymer Science 2012, 290, 261-274. Copyright © 2011, Springer-Verlag).

Hasegawa et al. studied the effect of uniaxial deformation on the orientation function of different dianhydride, diamine composition ratios in PI and PAA -

BPDA/PDA:ODA copolymer. In this study they found that the different diamine ratios in

PAA’s copolymer did not play a role in the chain orientation. All the different PAA compositions exhibited similar orientation function values throughout the stretching as can be seen inFigure 2.23. However, this observation was not valid anymore after imidization.

Increasing in the ODA ratios decreased the orientation to the MD during uniaxial stretching. [77]

FIGURE 2.23 ORIENTATION FUNCTIONS OF PAA(BPDA/PDA:ODA) AND PI(BPDA/PDA:ODA) WITH DIFFERENT COMPOSITIONS DURING UNIAXIAL STRETCHING. PDA:ODA RATIOS: (A) 70:30, (B) 50:50 AND (C) 30:70. [77] (adapted with permission from Spontaneous Molecular Orientation of Polyimides Induced by Thermal Imidization. 3. Component Chain Orientation in Binary Polyimide Blends” by M. Hasegawa, K. Okuda, M. Horimoto, Y. Shindo, R. Yokota, M. Kochi, Macromolecules 1997, 30, 5745-5752. Copyright © 1997 American Chemical Society).

Unsal et al. studied the real time dynamics and molecular mechanism of thermal imidization by using the URS-FTIR spectrometer. [73] In this study Unsal et al. cast PAA film which was constrained uniaxially to a frame. This study revealed that thermal imidization has five different stages as can be seen in Figure 2.24. First, during the increasing in temperature increments from RT to the imidization temperature, the bound solvent slightly evaporates from the system. As a result the PI matrix starts to shrink, which in turn results in increases in stress birefringence. In this stage the thermal imidization did not occur. The True stress and birefringence continue to increase. When the temperature reached to onset imidization temperature more solvent decomplexed from the PAA and evaporated or acted as a plasticizer in the polymer matrix. When the temperature increases to the temperature at which the imidization rate was fastest, the evaporation of trapped solvent increased. In the last step of the thermal imidization, the birefringence continued to increase due to formation of rigid polyimide chains and evaporation of solvent. [73]

FIGURE 2.24 REAL TIME MOLECULAR MECHANISM DURING THERMAL IMIDIZATION. [73] (adapted with permission from “Molecular mechanism of temporal physico/chemical changes that take place during imidization of polyamic acid: Coupled real-time rheo- optical and IR dichroism measurements” by E. Unsal, M. Cakmak. Polymer 2014, 55, 6569-6579. Copyright © 2014, Elsevier Ltd.).

CHAPTER III

REAL-TIME INFRARED−MECHANO-OPTICAL BEHAVIOR AND THE

STRUCTURAL EVOLUTION OF POLYPROPYLENE AND HYDROXYL-

FUNCTIONALIZED POLYPROPYLENE DURING UNIAXIAL DEFORMATION

Hydroxyl-functionalized polypropylene (PPOH) copolymer containing 0.4 mol% comonomer of 10-hydroxy-1-undecene was synthesized by the Zeigler-Natta copolymerization reaction between propylene and undecenyloxytrimethylsilane comonomers in order to add a polar group, which increases the material polarizability. As a result, the dielectric constant values of PPOH become higher than BOPP. Due to the high

PPOH dielectric constant, PPOH is a promising candidate for capacitor applications. The effect of hydroxyl groups on the thermal, structural and rheological properties of PPOH, as well as the crystallization kinetics and crystal morphology were studied before.

This chapter is focused on the effect of hydroxyl groups on structure evaluation and polymer chain segment behavior in the amorphous and crystalline domains during the processing cycle of heating, MD stretching, annealing, and cooling.

(This chapter is reprinted with permission from “Real-time Infrared-Mechano-

Optical Behavior and Structural Evolution of Polypropylene and Hydroxyl-Functionalized

Polypropylene during Uniaxial Deformation” by I. Offenbach, S. Gupta, T. C. M. Chung,

American Chemical Society.)

3.1 Introduction

Metallized biaxially oriented polypropylene (BOPP) represents the state-of-the-art in polymeric dielectric capacitors. [78,79] Despite having the highest breakdown strength among polymers, which is 600−750 V/μm for ∼10 μm thin films, [80,81] BOPP provides an energy density of only ~5 J/cm3 due to its low dielectric constant (∼2.2). One of the methods to achieve a high dielectric constant is to incorporate functional polar groups that align in the direction of applied electric field and impart an additional mechanism of ionic polarizability in the polymer. [80-82] Yuan et al. [81] recently showed that the incorporation of only 4 mol% hydroxyl groups in polypropylene (PP) results in a two-fold increase in the dielectric constant. Therefore, the hydroxyl-functionalized polypropylenes

(PPOH) are of considerable interest. Recent publications on the PPOH polymers have focused on the effect of hydroxyl groups on the thermal, structural and rheological properties of PP, [83] and the crystallization kinetics and crystal morphology, [84] all of which are important to consider during the melt-processing of PPOH polymers into thin films for capacitor application.

Thin polymeric capacitor films in the micrometer range are usually prepared by the tenter-frame biaxial stretching method. Tenter-frame biaxial stretching is a well- established technique [85,86] where semi-crystalline polymers, such as PP, are heated to a semi-molten state before stretching them in the machine direction (MD) followed by cooling and reheating and subsequent stretching in the transverse direction (TD).

Therefore, the polymers are subjected to a complex thermo-mechanical deformation during a processing cycle of the tenter-frame biaxial stretching method. [87] Moreover, the final state of polymer chain orientation and crystallinity is highly influenced by the processing conditions and dictate the final film properties and performance. While film tenter-frame biaxial stretching has been commercially employed for more than a decade, the structural evolution of polymer chains during such a process is still not well understood. Previously we had reported on the real-time structural development of PP during uniaxial deformation in the partially molten state. [58,88] In this paper, we report on the real-time structural development of a PPOH polymer, and focus on the behavior of polymer chain segments separately in the amorphous and crystalline domains during heating, MD stretching, annealing and cooling cycles of the tenter-frame biaxial stretching process. The results are compared with unmodified polypropylene.

3.2 Experimental Details

3.2.1 Materials

Capacitor-grade PP homopolymer (Borclean HB311BF) was donated by Borealis

(Mn = 53,000 g/mol, Mw = 227,000 g/mol, Mz = 529,000 g/mol, PDI = 4.2). PPOH polymer containing 0.4 mol% hydroxyl groups was synthesized by the Zeigler-Natta copolymerization reaction between propylene and undecenyloxytrimethylsilane comonomers, as shown in Figure 3.1. The synthesis procedure is described in detail elsewhere. [81] The reactivity ratios for this copolymerization were r1 = 68 for propylene and r2 = 0.032 for the comonomer, due to which the chemical structure of the PPOH polymers was rather blocky than random. [84]

FIGURE 3.1 SYNTHESIS OF PPOH POLYMER.

3.2.2 Sample Preparation

Unoriented polypropylene films were prepared in an extrusion casting line consisting of a Milacron extruder and a 12 wide slit die, casting rolls stack, six temperature– and motion–independent godet rolls, and a film collector. Table 3.1

summarizes the processing conditions along the casting line. A final thickness of 120 µm was obtained for the cast PP film. Unoriented PPOH films, 120 µm in thickness, were obtained by compression molding the polymer powder at 210°C in a TMP Vacuum

Compression Press. Dumbbell shaped specimens were cut from both cast and molded films for further characterization.

TABLE 3.1 PROCESSING CONDITIONS FOR PREPARING UNORIENTED POLYPROPYLENE FILMS.

Component Speed Temperature (°C) Zone 1 190 Zone 2 220 Single Screw Extruder 18.5 rpm Zone 3 220 Zone 4 225 Slit Die - - 220 1 96 Chiller Rolls 1.8 m/min 2 91 1 120 2 132 3 130 Godet Rolls 1.8 m/min 4 120 5 95 6 60

3.2.3 Characterization

PP and PPOH specimens were uniaxially stretched in the partially molten state at various temperatures to 100% strain at a stretch rate of 15 mm/min. A real-time mechano- optical measurement platform equipped with an ultra-rapid-scan polarized FTIR

spectrometer was used to probe the deformation behavior of polymer segments in the crystalline and amorphous domains separately. Details of the measurement technique are discussed elsewhere. [89] In brief, the specimen was clamped to load cells in the uniaxial stretching machine and enclosed within a thermal chamber. The stress-strain curves were calculated from the measured force and the true cross-sectional area of the deformed specimen, which was determined by measuring the real-time width of the specimen and assuming an incompressible deformation with transverse isotropy. The stress-strain data were further used to calculate the moduli of the PP and PPOH polymers. A visible wavelength light source and linear polarizers were used to measure the optical retardation

() through the polymer specimen. The birefringence (n) was calculated from the ratio of optical retardation () and the true specimen thickness.

A disk mirror interferometer was used to measure infrared (IR) absorbance at a rate of 300 spectra per second. The IR beam was split into two components, one parallel and another perpendicular to the machine direction (MD), which allowed the calculation of

Herman’s orientation function (fIR) using the following equation, [69]

(퐷−1)(퐷0+2) 푓퐼푅 = (3.1) (퐷+2)(퐷0−1) where D is the dichroic ratio, defined as the ratio of IR absorbance in parallel and perpendicular directions (A∥/ A⊥), and D0 is the dichroic ratio of an ideally oriented polymer, defined as

2 퐷0 = 2 cot 휓 (3.2)

where ψ is the transition moment angle of a specific side-group vibration with respect to the polymer chain axis.

The real-time mechano-optical measurements were made at four different processing steps in the following sequence – (1) preheating the polymer specimen from room temperature (RT) to the stretch temperature and one minute of holding, (2) isothermal uniaxial stretching, (3) isothermal annealing for 30 min. at the stretch temperature, and (4) cooling back to RT. The polymer specimen was confined between the clamps throughout these measurements.

The crystal structures were characterized with wide-angle X-ray diffraction

(WAXD) and small-angle X-ray scattering (SAXS) on films staked to provide sufficient thickness for good signal to noise ratio in data. WAXD measurements were performed using a Bruker AXS D8 Goniometer and SAXS measurements were performed using a

Rigaku MicroMax002+ diffractometer. Both instruments used CuKα radiation (λ = 0.1542 nm). SAXS was used to determine the long-spacing (LP) of the crystals. WAXD diffractograms were used to measure any changes in the crystal structure of PP upon stretching, and to compare the final values of orientation function with those obtained from

FTIR (fIR). The Herman’s orientation function was calculated from the WAXD data (fWAXD) using the following equation for an -monoclinic crystal structure of PP,

1 f = (3〈cos2 θ 〉 − 1) (3.3) WAXD 2 c,z

2 where θc,z is the angle between the c-axis and the stretch (z) direction (or MD). 〈cos θc,z〉 was calculated using the following equations, [58]

ퟐ ퟐ ퟐ 퐜퐨퐬 훉퐜,퐳 = ퟏ − ퟏ. ퟎퟗퟗ 퐜퐨퐬 훉ퟏퟏퟎ,퐳 − ퟎ. ퟗퟎퟏ퐜퐨퐬 훉ퟎퟒퟎ,퐳 (3.4)

ퟗퟎ° ퟐ ퟐ ∑ퟎ° 퐈퐡퐤퐥(훉)퐬퐢퐧훉퐜퐨퐬 훉 퐜퐨퐬 훉퐡퐤퐥,퐳 = ퟗퟎ° (3.5) ∑ퟎ° 퐈퐡퐤퐥(훉)퐬퐢퐧훉

where θhkl,z is the angle between the normal to the (hkl) lattice plane and the stretch direction, and 2 is the Bragg’s diffraction angle.

A TA Instruments differential scanning calorimeter (DSC Q-200) was used to measure the onset of melting (Tm,onset), melting point (Tm) and the heat of fusion (Hf) of

PP and PPOH films. Before the DSC measurements, temperature calibration was performed using an Indium standard (Tm = 156.6C). A heating rate of 10ºC/min was employed. The melting point was defined as the peak of the melting endotherm, with the largest peak denoted by Tm and the secondary peak denoted by Tmʹ. The melting onset temperature (Tm,onset) was measured at the point of intersection of the endotherm before the melting peak and the back-extrapolated curve from the endotherm after the melting peak.

The back-extrapolation was done by a straight line with the same slope, i.e. heat capacity, as the endotherm after the melting peak. The mass fraction crystallinity of the polymer (m) was calculated from the ratio of Hf/HPP, where Hf is the area under the DSC melting endotherm and HPP is the enthalpy of fusion of a 100% crystalline PP with an α- monoclinic structure, which is 165 J/g. [90]

In order to differentiate between the morphological changes that occurred in the polymers during each processing stage, DSC measurements were done separately on specimens that were sequentially removed from the uniaxial stretching chamber after each processing stage and subsequently cooled back to RT. These specimens are denoted as

(heat–cool), (heat–anneal–cool), (heat–stretch–cool) and (heat–stretch–anneal–cool) in the text below.

3.3 Results and Discussion

The thermal transitions for the PP and PPOH polymers are shown by the DSC endotherms in Figure 3.2. The addition of comonomer decreased the melting point (Tm) of

PPOH by 10C compared to PP, which is consistent with predictions from Flory’s theory of melting-point depression in copolymers. [91] The onset of melting (Tm,onset) also decreased for PPOH, by 15C compared to PP, which is attributed to the decreased crystallizability and formation of smaller polypropylene crystals in the PPOH copolymer.

[83]

66 J/g 80C 40% PPOH

130 C

T = 95C 155C m,onset PP Ts =141C

Hf =104 J/g m = 63%

Tm =165C

FIGURE 3.2 DSC CURVES FOR THE PP AND PPOH POLYMERS OBTAINED FROM THE

UNDEFORMED FILMS. THE SYMBOLS INDICATE THE FOLLOWING: TM,ONSET – ONSET OF

MELTING, TS – OPTIMUM STRETCHING TEMPERATURE DETERMINED FROM EQ. (6), TM –

MELTING POINT., ∆HF – HEAT OF FUSION, M – MASS FRACTION CRYSTALLINITY OF THE POLYMER.

An optimum temperature for stretching a semi-crystalline polymer in the semi- molten state (Ts) is typically determined using the relation, [58]

ퟏ 푻 = 푻 − (푻 − 푻 ) (3.6) 풔 풎 ퟑ 풎 풎,풐풏풔풆풕

The optimum stretching temperatures were calculated to be 141°C and 130°C for PP and

PPOH, respectively. In this study, we chose three deformation temperatures (Td) – Ts, Ts-

31, Ts+14 – to cover a broad range of the semi-molten regime based on the DSC exotherms in Figure 3.2.

3.3.1 Mechano-Optical Behavior

3.3.1.1 Heating to deformation temperatures.

Figure 3.3 shows the stress development and Figure 3.4 shows the birefringence development in PP and PPOH films during the 4 stages of processing. In the 1st stage, the polymer film was heated to the deformation temperature (Td) and isothermally maintained at that temperature for 1 minute. Prior to the heating, the specimen was slightly stressed to prevent surface wrinkling and sagging during heating. The change in specimen length before and after heating was measured to be negligibly small. All the polymer specimens were completely relaxed before uniaxial stretching in the 2nd stage, as indicated by a negligible stress at the deformation temperatures in Figure 3.3 (a). The average polymer chain orientation was isotropic before stretching as evidenced by the measured zero birefringence in Figure 3.4(a).

The FTIR absorption spectra of the PP and PPOH polymers (not shown) were quite similar except that PPOH showed two additional peaks at 3332 cm−1 and 3637 cm−1. [83]

The former corresponds to the stretching vibration of hydrogen bonded OH groups, while the latter corresponds to the stretching vibration of non-hydrogen bonded OH groups. The

H-bond strength decreased upon increasing the deformation temperature. [83,92] Figure

3.5 (a) and (b) show the real-time evolution of vertically and horizontally polarized IR absorptions in the crystalline and amorphous regions of PP and PPOH polymers, respectively, during each of the 4 processing stages. The dichroic ratios (D, Do) of crystalline and amorphous regions were calculated from the IR absorptions at 1045 cm-1 and 2720 cm-1, respectively. [75] The vibrational modes at 1045 cm-1 are C–C stretching and C–CH3 wagging with a crystalline transition dipole moment angle (c) of 0°. [93] The transition dipole moment angle in the amorphous regions (a) is 90°. The orientation functions (fIR) are shown in Figure 3.6 andFigure 3.7 for PP and PPOH, respectively, at temperatures corresponding to Ts-31 and Ts+14 C.

In the first heating stage, the orientation function was negligible for both amorphous

(fa) and crystalline (fc) phases, which is consistent with the calculated birefringence in

Figure 3.4(a).

20 (a) PP 110oC (b) (c) (d) 15 PP 141oC PP 155oC I II PPOH 99oC 10 PPOH 130oC 0 PPOH 144oC 0.02 5

True Stress (MPa) 0 60 80 100 120 140 0.0 0.2 0.4 0.6 0 5 10 15 20 25 30 140 120 100 80 60 40 Temperature (oC) Time (min) Temperature (oC) H

FIGURE 3.3 TRUE STRESS OF PP AND PPOH AT TS-31C, TS AND TS+14C DURING FOUR STAGES OF PROCESSING CYCLE: (A) HEATING + 1 MINUTE OF HOLDING , (B) STRETCHING, (C) 30 MIN OF ANNEALING, AND (D) COOLING. SOLID LINES REPRESENT PP WHILE THE DASH LINES REPRESENT PPOH. THE STRESS-STRAIN CURVE IN THE 2ND STAGE CAN BE DIVIDED INTO 2 REGIMES AS SHOWN FOR PP STRETCHED AT 110C, I: ELASTIC FLOW, AND II: YIELDING AND PLASTIC FLOW. THE INSET IN (B) SHOWS THE ELASTIC DEFORMATION REGION.

FIGURE 3.4. IN-PLANE BIREFRINGENCE OF PP AND PPOH AT TS-31C, TS, AND TS+14C DURING FOUR STAGES OF PROCESSING CYCLE: (A) HEATING + 1 MINUTE OF HOLDING, (B) STRETCHING, (C) 30 MIN OF ANNEALING, AND (D) COOLING. SOLID LINES REPRESENT PP WHILE THE DASH LINES REPRESENT PPOH.

(a) Crystalline peak 1045 cm-1 for PP 110°C

Amorphous peak 2721 cm-1 for PP 110°C

Crystalline peak 1045 cm-1 for PPOH 99°C (b)

Amorphous peak 2721 cm-1 for PPOH 99°C

FIGURE 3.5 REAL-TIME EVOLUTION OF IR ABSORBANCE FOR THE CRYSTALLINE PEAK AT 1045CM-1 AND THE AMORPHOUS PEAK AT 2721 CM-1 DURING THE 4 STAGES OF PROCESSING CYCLE (HEATING + 1 MIN. HOLDING, STRETCHING, ANNEALING AND COOLING) FOR (A) PP PROCESSED 110°C, AND (B) PPOH PROCESSED AT 99°C.

0.6 fa 110oC (a) (b) (c) (d) o 0.4 fc 110 C fa 155oC

IR o

f fc 155 C 0.2

0.0

40 60 80 100 120 140 0.0 0.2 0.4 0.6 0 5 10 15 20 25 30 140 120 100 80 60 40 o Temperature ( C) Time (min) Temperature (oC) H FIGURE 3.6 ORIENTATION FUNCTION OF AMORPHOUS AND CRYSTALLINE SEGMENTS IN PP DURING THE FOUR STAGES OF PROCESSING CYCLE: (A) HEATING + 1 MINUTE OF HOLDING, (B) STRETCHING, (C) 30 MINUTES OF ANNEALING, (D) AND COOLING. THE STARS REPRESENT THE CRYSTALLINE ORIENTATION FUNCTIONS OF THE FINAL SAMPLES AT ROOM TEMPERATURE, AND CALCULATED FROM OFFLINE WAXD MEASUREMENTS USING EQUATION (3).

0.6 fa 99oC (a) (b) (c) (d) fc 99oC 0.4 fa 144oC fc 144oC

IR f 0.2

0.0

40 60 80 100 120 1400.0 0.2 0.4 0.6 0 5 10 15 20 25 30140 120 100 80 60 40 o Temperature ( C) Time (min) Temperature (oC) H FIGURE 3.7 ORIENTATION FUNCTION OF THE AMORPHOUS AND CRYSTALLINE SEGMENTS IN PPOH DURING THE FOUR STAGES OF PROCESSING CYCLE: (A) HEATING + 1 MINUTE OF HOLDING, (B) STRETCHING, (C) 30 MINUTES OF ANNEALING, (D) AND COOLING. THE STARS REPRESENT THE CRYSTALLINE ORIENTATION FUNCTIONS OF THE FINAL SAMPLES AT ROOM TEMPERATURE, AND CALCULATED FROM OFFLINE WAXD MEASUREMENTS USING EQUATION (3).

Although not much effect was observed on the polymer chain orientation upon heating the specimen to Td and isothermally maintaining for a minute, changes in the crystal morphology occurred for both PP and PPOH as indicated by measurements of the crystallinity and the melting point. DSC exotherms measured on the heat–cool specimens are shown in Figure 3.8, and the values of m and the melting points are summarized in

Table 3.2.

An existence of a second endothermic peak or shoulder (Tmʹ) for heat–cool specimens indicated an increase in the crystal population of that size which would melt at the corresponding Tmʹ. Such an increase is attributed to lamella-thickening which involves simultaneous melting of the smaller crystals (with Tm ≤ Td) and recrystallization of mobile amorphous segments onto the bigger crystals. Consequently, the onset of melting also increased with an increase in the deformation temperature, as shown in Figures 3.8, and

3.9(a), although the melting point, Tm, remained unchanged. The lamella-thickening was also accompanied by an increase in the total crystallinity of heat-cool specimens compared to the unprocessed films. However, the exact state of crystallinity at Td before cooling is unknown as it temporally changes.

The extent of lamella-thickening, with (Tm - Tmʹ) as its measure, increased with the deformation temperature for both PP and PPOH, i.e. (Tm - Tmʹ) decreased at higher Td as shown in Figure 3.9(b). However, it was more pronounced in PP than PPOH at Ts or relative temperatures. Similar observations were made on the heat–anneal–cool polymer specimens that were heated to the deformation temperature and isothermally annealed for

30 min. before cooling. Both the melting onset and the secondary melting temperature

increased with the annealing temperature for PP and PPOH, with the extent of lamella- thickening more pronounced in PP as shown in Figure 3.9 (b).

Tm,onset

Heat FlowHeat (a.u) Tmʹ

PP 110oC Tm PPOH 99oC o PP 141 C PPOH 130oC PP 155oC (a) PPOH 144oC (b)

60 80 100 120 140 160 180 60 80 100 120 140 160 180 o o Temperature ( C) Temperature ( C) FIGURE 3.8 DSC ENDOTHERMS FOR (A) PP AND (B) PPOH AFTER HEATING TO THE DEFORMATION TEMPERATURE, HOLDING FOR A MINUTE AND COOLING BACK TO RT (HEAT–COOL). Heat-Cool 160 60 Heat-Cool Heat-Anneal-Cool (a) (b) Heat-Anneal-Cool 150 Heat-Stretch-Anneal-Cool 50 Heat-Stretch-Anneal-Cool 140

C) 130 40



( 

120 (

ʹ 30

110 m

T -

m,onset 100 20

m T 90 PP T 10 80 PPOH 70 0 -40 -30 -20 -10 0 10 20 -40 -30 -20 -10 0 10 20 Td-Ts (C) Td-Ts (C)

FIGURE 3.9 (A) ONSET OF MELTING, AND (B) DIFFERENCE BETWEEN THE PRIMARY AND SECONDARY MELTING TEMPERATURES AS A MEASURE OF LAMELLA-THICKENING, FOR PP (FILLED SYMBOLS, DASHED LINES) AND PPOH (OPEN SYMBOLS, DOTTED LINES) POLYMERS SUBJECTED TO DIFFERENT DEFORMATION TEMPERATURES. THE LINES ARE DRAWN FOR CLARITY.

TABLE 3.2 THERMAL PROPERTIES OF PP AND PPOH POLYMERS BEFORE AND AFTER EACH PROCESSING STAGE. Polymer Unprocessed Deformation Heat–Stretch– Heat–Cool Heat–Anneal–Cool Heat–Stretch–Cool Film Temperature Anneal–Cool

m Tm,onset Tmʹ Tm Td m Tm,onset Tmʹ Tm m Tm,onset Tmʹ Tm m Tm,onset Tmʹ Tm m Tm,onset Tmʹ Tm

(%) (C) (C) (C) (C) (%) (C) (C) (C) (%) (C) (C) (C) (%) (C) (C) (C) (%) (C) (C) (C)

Ts-31 110 66 103 125 164 67 106 133 165 48 122 155 164 61 114 - 163

PP 63 95 - 165 Ts 141 67 108 150 164 63 112 144 164 50 136 161 165 68 123 - 163

Ts+14 155 68 115 - 165 65 116 162 165 58 148 163 165 69 127 - 164

Ts-31 99 42 82 105 155 44 92 104 158 39 93 110 157 43 86 103 157

PPOH 40 80 - 155 Ts 130 47 96 133 156 38 98 130 157 30 108 145 156 49 106 132 157

Ts+14 144 45 96 149 157 39 110 145 157 34 119 149 157 53 110 147 157

The measurement errors in crystallinity (m) are ±2%, melting onset (Tm,onset) are ±2C, and Tmʹ and Tm are ±1C.

A direct evidence of lamella thickening was obtained from the SAXS measurements. Figure 3.10(a) shows the plots of the SAXS intensity (Iq2 vs q, where q is the scattering wave vector) for heat-anneal-cool specimens of PP. The inset is the 2D SAXS profile for the heat-anneal-cool specimen of PP at 110C which shows that the lamellar crystals were isotropically oriented. The long-spacing of lamellar crystals (LP) was calculated using the following equation,

LP = 2/qpeak (3.7)

where qpeak is the peak position in Figure 3.10(a). The values of LP are plotted in Figure

3.10(b). For the heat-anneal-cool specimens, it was observed that the long-spacing was higher compared to the unprocessed film, increased with an increase in the annealing temperature, and was higher for PP compare to PPOH. These observations are consistent with the DSC measurements of the melting onset and the secondary melting point.

Heat-Anneal-Cool (a) PP 30 Heat-Stretch-Cool 155C Heat-Stretch-Anneal-Cool 25

(b)

(nm)

(a.u.) P

2 20 L

I.q 141C 15 PP

110C qpeak PPOH 10 0 0.02 0.04 0.06 0.08 0.1 0.12 -40 -30 -20 -10 0 10 20 -1 q (A ) Td-Ts (C) FIGURE 3.10 (A) SAXS INTENSITY PROFILE AS A FUNCTION OF THE SCATTERING WAVE VECTOR. SOLID CURVES CORRESPOND TO HEAT-ANNEAL-COOL SPECIMENS OF PP. DOTTED CURVES CORRESPOND TO SPECIMENS THAT WERE STRETCHED UNTIL THE POINT OF MINIMUM BIREFRINGENCE IN FIGURE 3.4(B) AND COOLED. THE CURVES ARE SHIFTED

VERTICALLY FOR CLARITY. THE INSET IS A 2D PROFILE FOR THE HEAT-ANNEAL-COOL SPECIMEN OF PP AT 110C INDICATING RANDOM ORIENTATION. (B) LONG-SPACING OF LAMELLAR CRYSTALS OF PP (FILLED SYMBOLS, DASHED LINES) AND PPOH (OPEN SYMBOLS, DOTTED LINES) SUBJECTED TO DIFFERENT DEFORMATION TEMPERATURES. THE LINES ARE DRAWN FOR CLARITY.THE LONG-SPACING OF UNDERFORMED PP AND PPOH IS SHOWN BY THE HORIZONTAL DASHED LINES.

One of the possible reasons for a reduced lamellar thickening in PPOH polymer could be the steric hindrance from the copolymer side-chains and the pendant hydroxyl groups that prevents periodic folding of the polypropylene backbone chains into a crystal structure. Interestingly, certain differences between the heat–cool and heat–anneal–cool specimens were observed. For example, in Figure 3.9(b), the extent of lamella thickening decreased for the heat–anneal–cool specimens compared to the heat–cool specimens which indicated that melting became more pronounced than recrystallization upon prolonged exposure at the deformation temperatures.

3.3.2 Uniaxial Stretching

While not much difference was observed between PP and PPOH in the 1st heating stage in Figure 3.3 (a), 3. 4(a), 3. 6(a) and 3. 7(a), it was quite evident in the 2nd stage where the polymer specimens were deformed uniaxially. The true stress–true (Hencky) strain curves at all deformation temperatures are shown in Figure 3.3(b), and the calculated moduli data are given in Table 3.3. A typical behavior characteristic of a semi-crystalline polymer was observed with an elastic deformation at lower strains (marked as regime I in Figure

3.3(b)) followed by yielding and necking (marked as regime II and separated by a dashed

vertical line from regime I). The Young’s modulus (E) and the yield strength (y) naturally decreased upon increasing the deformation temperature.

The contributions to the deformation stress in a semi-molten state come from both the crystalline network and the trapped entanglements. The average entanglement spacing, represented by the tube diameter (a) in polymer reptation tube model, decreases upon decreasing the temperature. [94] Assuming that the incorporation of 0.4 mol% does not affect the entanglement spacing in PP, at the reference temperature of Ts – 풂퐏퐏(ퟏퟒퟏ퐂) >

풂퐏퐏퐎퐇(ퟏퟑퟎ퐂). This implies that the trapped entanglement contribution to the stress should be larger in PPOH than in PP at Ts. In addition, the PPOH polymer exhibited a higher melt elasticity and viscosity than PP [92] due to intermolecular hydrogen bonding interactions between the pendant OH groups, which is expected to require a further increase in the deformation stress in PPOH for a given strain. However, as observed in Figure 3.3(b) and Table 3.3, both the Young’s modulus and the yield strength were considerably lower for PPOH compared to PP at Ts or relative temperatures. Furthermore, yielding, which is associated with the unraveling of polymer chains from the lamellar crystals, occurred at lower strains in PP, see Table 3.3. These observations indicate that the material response in the semi-molten state was dominated by the remaining crystalline network over that of

H-bonding between the OH groups or the trapped entanglements, and is consistent with a lower crystallinity of PPOH.

TABLE 3.3 MECHANICAL PROPERTIES OF PP AND PPOH POLYMERS. Yield Deformation Young's Yield Photo-elastic Stress Polymer (Hencky) Temperature Modulus Strength Coefficients Relaxed Strain 3 -1 Td E y y Cʹo×10 (MPa ) rel

(C) (MPa) (MPa) I II  IV (MPa)

Ts-31 110 358.2 8.8 0.06 -0.11 -0.63 6.18 1.48 9.2

PP Ts 141 150.4 3.4 0.05 -0.16 -0.93 11.95 2.17 3.5

Ts+13 155 117.5 1.8 0.05 -0.27 -5.97 5.84 1.61 2.2

Ts-31 99 70.6 3.8 0.08 -0.02 1.03 2.87 1.82 3.4

PPOH Ts 130 10.0 1.4 0.13 0.28 3.02 4.88 2.61 1.1

Ts+13 144 4.5 0.5 0.13 0.50 3.22 7.12 - 0.6

The differences in structural evolution between PP and PPOH during uniaxial stretching were also evident from the birefringence and FTIR measurements. The strain- optical curves shown in Figure 3.4(b) indicate that the orientation development at Ts or relative temperatures was higher in PP than PPOH, similar to the stress development. The average polymer chain orientation, with birefringence (n) as its measure, can be related to the crystalline and amorphous segmental orientations (fc, fa) and crystallinity (m) using the two phase model,

ퟎ ퟎ 횫풏 = 풇풄흓풎횫풏풄 + 풇풂(ퟏ − 흓풎)횫풏풂 + 횫풏풇풐풓풎 (3.8)

0 0 where nc and na are the intrinsic birefringence of the polymer chains in the crystalline and amorphous phases, respectively, and nform is the form-birefringence which is assumed zero in the present case. Due to the exclusion of OH groups and comonomer side-chains

0 from the crystalline phase in the PPOH polymer, [83] nc was considered to be equal for

0 PP and PPOH. Also, considering the small comonomer content in the PPOH polymer, na was assumed to be equal for PP and PPOH.

The birefringence data in Figure 3.4(b) were supplemented by the orientation functions of crystalline and amorphous phase, which are shown in Figure 3.6(b) and Figure

3.7(b) for PP and PPOH, respectively. Both the amorphous and crystalline segment orientation gradually increased in the direction of deformation from an unoriented state, which is consistent with the birefringence measurements. However, a few differences were observed between PP and PPOH. Firstly, a slightly negative birefringence was measured in PP up to a Hencky strain of ~0.16, see Figure 3.4(b), during which the polymer had already transitioned from an elastic to a plastic deformation regime. Afterwards the birefringence gradually increased to zero and became positive. The magnitude of negative birefringence was highest at the lowest deformation temperature. This observation is supported independently by the data in Figure 3.6(b) where both the amorphous and crystalline orientation functions become slightly negative at the beginning of deformation before increasing to a positive value.

The individual contributions from the crystalline and amorphous phases to the overall birefringence during uniaxial stretching of polymers have been previously reported.

Samuels [75] observed that the orientation function of the amorphous phase during uniaxial stretching of PP was negative, i.e. amorphous PP chains were oriented in the TD, at strains up to 100% beyond which it became positive. This was found true for both compression- molded and cast-and-annealed samples. However, unlike the results reported in this study, the crystalline orientation functions and the overall birefringence were measured to be

always positive. On the contrary, Sasaguri et al. [95] showed that PP exhibits a negative birefringence upon uniaxially stretching at low strains after ageing for several days. This is attributed to the negative contributions from the crystalline phase. Opposite to the observations by Samuels and Sasaguri et al., other studies [96,97] on uniaxial stretching of

PP show a positive amorphous and crystalline orientation functions.

The results obtained in Figure 3.4(b), 3.6(b) and 3.7(b) indicate that the spherulitic deformation in PP at the beginning of uniaxial extension results in a morphology where a fraction of the individual lamellae become aligned towards the MD with crystalline segments comprising the lamellae oriented towards the TD. One of the mechanisms through which this is possible is the lamellar bending, twisting and splaying within the equatorial planes of spherulites followed by the lamellar break-up. Similar models have been proposed for polyethylene deformation by Shimamura et al. [98] and Michler et al.

[99] based on SEM and TEM observations, respectively, and for polybutene-1 deformation by Sasaguri et al. [100] As a consequence of this morphology, the inter-lamellae tie chains in the amorphous phase may also be forced to orient in the TD. Upon further deformation of the polymer specimen beyond the elastic regime, the amorphous tie chains and those unraveled from the crystalline regions during necking gradually orient in the MD. This is simultaneously followed by further lamellar break-up and their rotation in TD along their length, which results in the crystalline segments comprising the lamellae to orient in the

MD. The above result is also consistent with AFM observations of crystalline morphology in a polyamide film subjected to uniaxial deformation. [101]

Contrary to PP, such an observation was not made for the PPOH polymer in Figure

3.4(b) orFigure 3.7(b). It appears that the crystal lamellae in PPOH bear the initial load

during uniaxial deformation. This was evident from a zero fa but a non-zero positive fc in

Figure 3.7(b) during the initial deformation. A non-zero positive fc indicates that the spherulitic break-up in PPOH commenced by rotation of the lamellae in the TD which resulted in the MD orientation of crystalline segments. The amorphous tie-chains were negligibly affected during this process.

Although the reasons for this different behavior of PP and PPOH are not clear, one of the plausible reasons could be the coarser spherulitic morphology of the PPOH polymer with lower crystallinity. [84]

The stress-optical curves for the stretching stage are shown in Figure 3.11 for PP and PPOH. Analogous curves for the orientation function vs. stress for PP and PPOH are shown in Figure 3.12(a) and (b), respectively. In general, four regimes with different photoelastic coefficients (Cʹo) could be distinguished, as marked in Figure 3.11 for PP stretched

st at 110C. The values of Cʹo are given in Table 3.3. In the 1 regime, which is characterized by elastic deformations at low strains in Figure 3.3(b), birefringence in PP became negative and that in PPOH became positive (except at 99C where it was negligibly small) indicating different mechanisms of spherulitic break-up, as discussed previously. As evident from

Figure 3.12(a) and (b), this is caused by a negative fc and a negative fa for PP but a positive

nd fc and zero fa for PPOH. In the 2 regime which is characterized by a transition from elastic deformation to yielding and commencement of necking in Figure 3.3(b), Cʹo further decreased for PP and increased for PPOH indicating further growth of the bulk morphology observed in regime I.

0.014 I II III IV 0.012 0.010 0.008 PP 110oC 0.006 PP 141oC 0.004 PP 155oC o 0.002 PPOH 99 C Birefringence PPOH 130oC 0.000 PPOH 144oC -0.002 0 5 10 15 20

True Stress (MPa)

FIGURE 3.11 STRESS-OPTICAL CURVES FOR PP AND PPOH POLYMERS AT DIFFERENT DEFORMATION TEMPERATURES FOR THE STRETCHING STAGE. DIFFERENT REGIMES FOR PP STRETCHED AT 110C ARE MARKED AND SEPARATED BY THE DOTTED LINES.

FIGURE 3.12 ORIENTATION FUNCTIONS FOR CRYSTALLINE AND AMORPHOUS POLYMER CHAIN SEGMENTS AS A FUNCTION OF TRUE STRESS FOR (A) PP AND (B) PPOH AT DIFFERENT DEFORMATION TEMPERATURES. DIFFERENT REGIMES CORRESPONDING TO FIGURE 11 ARE MARKED AND SEPARATED BY THE DOTTED LINES FOR (A) PP STRETCHED AT 110C, AND (B) PPOH STRETCHED AT 99C. THE INSET IS A MAGNIFICATION OF THE SELECTED AREA ON THE PLOT.

The 3rd and 4th regimes in Figure 3.11 were characterized by neck propagation in

Figure 3.3(b). In the 3rd regime, a sudden increase in birefringence was observed for both

PP and PPOH. The presence of such a “stress–offset" in the development of optical anisotropy is consistent with previous observations in semi-crystalline [58,102] polymers stretched in the semi-molten regime as well as amorphous [103] polymers stretched near the Tg. As evident from Figure 3.12(a) and (b), the rate of orientation development in this regime was almost equal for both amorphous and crystalline segments in PP, but was

th higher for the amorphous than crystalline segments in PPOH. In the 4 regime, Cʹo decreased slightly compared to the 3rd regime which could be attributed to the gradual saturation in polymer chain orientation as they approach their limit of extensibility.

An interesting observation from Figure 3.11 is that the stress–offset for birefringence development decreased upon increasing the deformation temperature for both PP and PPOH, and was higher for PP than PPOH at relative deformation temperatures.

However, the strain–offset (calculated from Figure 3.3(b) as the strain at the stress-offset) varied only slightly between PP and PPOH, extended into the plastic deformation regime, and was independent of the deformation temperature (Figure 3.4 (b)). This universality of the strain–offset suggests that the orientation of polymer chains in the direction of deformation was mainly restricted by the process of yielding, and that the stress merely scaled inversely with the temperature.

3.3.3 Thermal annealing and cooling.

The uniaxial deformation was followed by thermal annealing in the 3rd stage where the stretched specimen was kept fixed between the clamps at the stretching temperature for

30 min. Even in this stage, differences in the structural evolution between PP and PPOH were observed. Figure 3.3(c) shows the stress evolution Figure 3.4 (c) shows the change in birefringence, and Figure 3.6(c) and Figure 3.7Figure 3.7(c) show the change in orientation functions for PP and PPOH, respectively, as a function of annealing time. Stress relaxation was observed in all cases; however the stress did not relax back to zero but leveled off at an arbitrary value. The value of the total stress relieved upon annealing (rel) is given in

Table 3.2. Interestingly, rel was very close to the yield stress (y) suggesting that the major contributions to the stress relaxation come from the “elastic” motions and only minor contributions may come from the large-scale segmental motion or reptation of the polymer chains.

The birefringence increased only slightly upon annealing, which was consistent with the orientation function measurements. For PP, the crystalline orientation function decreased and the amorphous orientation function increased for early annealing times up to 3 min. before stabilizing to a constant value and, hence, no significant change in the overall birefringence. The occurrence of thermal crystallization at the annealing temperature would be expected to produce MD-oriented crystalline segments due to MD- oriented amorphous chains thereby increasing the fc, which is not observed in Figure 3.6(c).

Also, in such a case, one would expect the relaxation time-scale to be temperature-

dependent which is again not observed from Figure 3.6(c). One of the possibilities for a decrease in fc could be the relaxation resulting from the minor spatial rotation of oriented lamellar crystals as a mechanism of stress release. The amorphous tie chains that are trapped between the crystalline junctions can only undergo limited reptation but can relieve stresses by axial rotation of the strained bonds. The spatial rotation of the lamellar crystals and the strained bonds is also consistent with the “elastic” nature of stress relaxation and, combined with limited reptation of amorphous chains, may result in rel > y.

In contrast to PP, smaller changes in the amorphous orientation function or the crystalline orientation function were observed for PPOH in Figure 3.7(c), and rel < y from Table 3.2(except PPOH annealed at 144C) indicating that the elastic stress imposed during uniaxial deformation was not fully recovered. Both these results can be attributed to a more complex semi-crystalline morphology of the PPOH polymer compared to PP due to the presence of H-bonding, [84] which increases the viscosity and inhibits segmental motion of the polymer chains.

Cooling the stretched and annealed polymer specimen from deformation temperatures back to room temperature in the 4th stage led to an increase in the stress, Figure

3.3(d), primarily due to thermal shrinkage of the specimens which were still spatially confined between the clamps in the machine direction but free to contract in transverse directions. For both polymers, the thermal stresses were higher but the percent increase was lower upon cooling from a lower deformation temperature. The increase in stress was larger for PP than PPOH, which can be attributed to the higher crystallinity of PP analogous to creating more physical cross-linked joints inside the polymer. Birefringence measurements in Figure 3.4(d) showed that the polymer chain orientation did not change

much upon cooling, which is consistent with a negligible change in the orientation functions, fa and fc, in Figure 3.6(d) and Figure 3.7(d) for PP and PPOH, respectively.

Stretching and annealing the polymer specimen at different deformation temperatures also resulted in a change in the crystallinity and the secondary melting point, see Table 3.2(heat–stretch–anneal–cool). In order to decouple the effect of stretching from annealing on m and Tmʹ, DSC measurements were performed on the specimens that were cooled immediately after stretching. These values are given in Table 3.2 (heat–stretch– cool). For the heat–stretch–cool specimens, the crystallinity was higher than the unprocessed film and increased with an increase in the deformation temperature for both

PP and PPOH. Since stretching up to high strains is expected to break lamellar crystals and reduce crystallinity, an increase in the crystallinity of both PP and PPOH was surprising.

In conjunction with the crystallinity, the long-spacing also increased with the deformation temperature, refer to Figure 10(b), with simultaneous increases in Tm,onset and Tmʹ in Table

3.2. In light of this observation it is reasonable to postulate that stretching of the polymer specimen breaks the smaller crystals first before breaking the larger ones. When the stretching is stopped and the polymer cooled, thermal crystallization occurs where the molten polymer chains crystallize on the existing lamellar crystals. Cooling from a higher temperature provides more time for polymer chains to crystallize, hence m, LP, Tm,onset and

Tmʹ increase with an increase in the deformation temperature. Clearly, thermodynamically driven lamella thickening occurs at the expense of smaller crystals. It is interesting to note that the average long-spacings in the PP specimens that were stretched up to the point of lowest birefringence and cooled were the same as those for the heat-anneal-cool specimens, see Figure 3.10(a).

Similar lamella thickening was observed for the heat–stretch–anneal–cool specimens as well. Their crystallinity was lower than the heat–stretch–cool specimens, which is consistent with previous observations of a lower crystallinity of heat–anneal–cool specimens compared to heat–cool specimens (refer to Table 3.2). Both of these results suggest that melting is more pronounced than recrystallization upon prolonged annealing at the deformation temperatures. The total crystallinity of heat–stretch–anneal–cool specimens of PP increased with an increase in Td whereas that for PPOH decreased, but it was lower than the unprocessed film for both the polymers. A decrease in m with an increase in Td for PPOH was also observed for the heat-anneal-cool specimens, and can be attributed to the steric hindrance by the copolymer side-chains and the pendant OH groups to the regular folding of PP chains into a crystal structure while the polymer was cooling back to room temperature.

3.3.4 Crystal Structure

The crystalline orientation function calculated from FTIR measurements was confirmed independently with WAXD measurements. Figure 3.13 shows the WAXD diffractograms for PP and PPOH at the end of the processing cycle. No change was observed in the crystal structure of both polymers, which was an -monoclinic polypropylene crystal. [104-105] The Hermans orientation function (fWAXD) was calculated using equations (3)–(5) based on the (110) and (040) lattice planes. The calculated values are compared with fIR in Figure 3.6(d) and Figure 3.7(d), and were fairly close for PPOH but higher for PP.

1200 (110) (040) PP 110oC 1000 14.2 17.1 PPOH 99oC

800

600

Intensity 400

200

0 5 10 15 20 25 30 35 2 theta FIGURE 3.13 WIDE-ANGLE X-RAY DIFFRACTOGRAMS OF PP AND PPOH POLYMERS AFTER A FULL PROCESSING CYCLE (HEATING + 1 MIN HOLDING, STRETCHING, ANNEALING AND COOLING). THE INSET SHOWS A 2D WAXD PATTERN OF PP STRETCHED AT 110C.

3.4 Structural Evolution

Based on the above results and discussions, a schematic is provided in Figure 3.14 for the structural evolution of polypropylene homopolymer during uniaxial stretching in the partially molten state focusing on why we observe negative birefringence during real time stretching of PP. Within the partially molten regime (i.e. between Tm,onset and Tm), both the melting and recrystallization process are occurring dynamically. Based on the previous [98-101] and current results, the deformation of spherulitic structures tends to start in the equatorial regions of the spherulites. Lamella twisting occurs first, followed by the lamella break up and rotation. This latter mechanism leads to crystalline chains in the lamellae orient perpendicular to the stretching direction. Since the PP chains exhibit

o o positive intrinsic birefringence (nc =0.0331, na =0.0468), [106] initially optically isotropic material temporarily exhibits negative birefringence (nTD > nMD). With further increase in deformation, the chains unraveling from the lamellae begin to more substantially orient in the machine direction to eventually reverse this trend and the birefringence becomes positive.

Deformation zone

Deformation Lamella and twist break-up

Undeformed c-axis in c-axis in MD TD

Gradual c-axis 0 orientation in MD

εH

FIGURE 3.14 SCHEMATIC DEPICTING THE STRUCTURAL EVOLUTION OF POLYPROPYLENE DURING UNIAXIAL STRETCHING FROM THE PARTIALLY MOLTEN STATE.

3.6 Conclusions

The structural evolution of polypropylene during heating, uniaxial stretching, annealing and cooling was studied, and the effect of incorporation of 0.4 mol% comonomer of 10-hydroxy-1-undecene in PP was investigated. The mechano-optical behavior was primarily dominated by the crystalline morphology, but was influenced by the presence of intermolecular H-bonding and the C9 side chains of the copolymer. Independent and simultaneous measurements of the in-plane birefringence and orientation functions of the amorphous and crystalline segments shed light on the behavior of the polymer segments during each stage of the deformation. The break-up of spherulite morphology at the early stages of uniaxial deformation was found to be different for PP as compared to PPOH. In

PP, the spherulitic break-up led to break up and rotation of lamellae fragments in the equatorial regions of the spherulites first, with this rotation leading to observation of a negative birefringence. Further stretching reversed this trend and thus birefringence sign become positive as the amorphous and crystalline chains attained increasingly large preferential orientation in the MD. This initial negative birefringence was not observed in

PPOH. Four different regimes in the stress-optical curve for uniaxial deformation were observed for both polymers and explained on the basis of changes in crystal morphology.

An increase in the polymer chain orientation in the MD was always preceded by a stress- offset that decreased with an increase in the deformation temperature. Growth of a second crystal population and lamella thickening was observed in most cases for both PP and

PPOH polymers. However, the crystal size was always lower for PPOH due to steric hindrance by the hydroxyl groups and the copolymer side-chains.

CHAPTER IV

REAL-TIME MECHANO-OPTICAL BEHAVIOR AND STRUCTURAL EVOLUTION

OF POLYIMIDE (BTDA-DAH) DURING UNIAXIAL DEFORMATION.

This chapter focuses on the real-time mechano-optical behavior of PI(BTDA-

DAH). The mechano-optical behavior of the polymer was studied in its glassy and rubbery states as a function of processing temperature and stretching rate during uniaxial deformation. This study was performed with an instrumented system that combined real- time spectral birefringence measurements with true stress and true strain during uniaxial deformation. Three regimes of stress optical behavior were revealed. First, at the early stage of deformation the stress optical rule was observed; birefringence linearly increased with a stress optical constant of 17.8 GPa-1: regime I. Second, a deviation from linearity took place. At a higher temperature and/or lower stretching rate the deviation was positive and the birefringence rapidly increased while the stress slowly increased: regime II. At a lower temperature and/or higher stretching rate, this deviation from linearity was negative: regime IIIa. Third, in cases where regime II was revealed, it was followed by a negative deviation of birefringence from linearity and reached a plateau while stress rapidly increased: regime IIIc.

According to off-line characterization techniques--differential scanning calorimetry and wide-angle X-ray diffraction--the material remained amorphous during

regime I and the early stage of regime II. By the end of regime II, a rapid increased in the crystallinity was observed. This implies stress induced crystallization associated with regime II. There was no significant change in the crystallinity with further stretching into regime III, where the polymer chains reached their finite extensibilities.

4.1 INTRODUCTION

There has been a recent surge in research on the energy density of polymeric dielectric capacitors; this surge is due primarily to an increased need for energy storage devices which charge quickly and have high energy release rates. [3,107-109] High energy density, fast charging times, and high energy release rates are crucial parameters for novel energy storage systems technology such as electromagnetic armor, [78,111] electromagnetic rail guns, [78] hybrid cars, [110] particle beam accelerators, [112] and others. [2,78,111,112]

Metallized biaxially oriented polypropylene (BOPP) is often considered the most reliable and pervasive for today’s polymeric dielectric capacitors application. [78,79] It has a low loss factor (<0.02%), [79] dielectric constant of 2.2, [79-81] a dielectric strength of

~720 [V/ μm], [80] and has self-healing characteristics. [79] However, it is limited by an energy density of 5 (J/cm3) at breakdown, [80] and the breakdown strength decreases at temperatures higher than 85°C. Furthermore, the operating temperature is limited to 105℃.

[113] Therefore capacitors which are designated to operate at temperatures higher than

105℃ are made from different polymers such as Poly(ethylene napthalate), [79,113]

Poly(phenylene sulfide), [79,113] Polyetherimide, [114] etc. However, these materials 81

usually suffer from high dielectric loss, which is highly undesirable for capacitor applications. Polyimides are a class of high performance polymers due to their high heat resistance as well as their chemical, mechanical, and electrical properties. [39-41] In the past, research on dielectric behavior of polyimides was focused on microelectronics applications; this research attempted to synthesize a polyimide with a low dielectric constant. [40,41] However, researchers have lately investigated the possibility of using

Polyimide as a polymeric dielectric capacitor for high-temperature applications because of its highly heat resistant nature. [115-118] Rui et al. synthesized a series of novel Polyimide polymers that have dielectric constants up to 7.8 with low loss factors (<1%). [119]

Therefore, these polymers can be used to create capacitors. One of those promising materials was PI(BTDA-DAH), which exhibited a dielectric constant of 3.57 and a breakdown strength of 812 [V/ μm]. [119]

The conditions under which thin polymeric films are manufactured impacts the final state of their polymer chain orientation and crystallinity; in turn, these factors largely dictate the dielectric properties of the polymeric films. For example, unoriented PP films exhibited spherulitic structures. The larger the size of the spherulitic structure, the lower the film’s maximum breakdown strength. [120] When stretched, Polypropylene’s spherulitic structure converted to a fibrillar one; [88,121] this increased its breakdown strength. Guan et al. studied the effect of polymer chain orientation on polarizability in

P(VDF-HFP). They found that stretching P(VDF-HFP) films increased their polymer polarizability because it resulted in a higher distribution of the dipole moments of CF2 in

PVDF crystals parallel to the electric field. [122] Due to the difficulty of monitoring fast

changes in the structural development of polymers during deformation, the observance of such changes while preparing capacitor films is not often undertaken.

This chapter focuses on the real-time structural development of a PI(BTDA-DAH), and particularly on the structural evolution during uniaxial deformation as it plays a critical role in the electrical and mechanical performance of these materials.

4.2 Experimental Details

4.2.1 Materials

PI(BTDA-DAH) polymer was synthesized by a two-step polycondensation reaction shown in Figure 4.31. First, Poly(amic acid) was synthesized by a reaction between NMP (acting as the solvent), 3,3′,4,4′-benzophenone tetracarboxylic dianhydride

(BTDA) and 1,6-diaminohexane (DAH) monomers with isopropylamine acting as the end- capping reagent. The reaction was carried out with the N-methyl pyrillodinone (NMP) as the solvent. In the second step, PI(BTDA-DAH) was formed by the imidization of polyamic acid. The synthesis procedure is described in detail elsewhere. [119]

NMP

R.T

NMP

170°C -180°C

FIGURE 4. 1 SYNTHESIS OF PI (BTDA-DAH) POLYMER. [119]

4.2.2 Sample Preparation

Thin films of PI(BTDA-DAH) were prepared by solution casting from a 11wt% solution in M-cresol. Before the casting process, the solution was filtrated through a 1 µm

PTFE filter. The films were cast on a glass substrate by using a 6'' wide casting doctor blade. The films were dried at 80°C for 8 hours and a final thickness of 20 µm was obtained.

Dumbbell-shaped specimens were cut out from the film for further characterization.

4.2.3 Characterization

PI(BTDA-DAH) specimens were uniaxially stretched at glassy and rubbery states at various temperatures and rates. An instrumented real-time mechano-optical measurement platform was used to investigate the deformation behavior of PI(BTDA-

DAH) coupled with birefringence. Details of the measurement technique are discussed elsewhere. [58] In brief, the specimen was clamped to load cells in the uniaxial stretching machine and enclosed within a thermal chamber. The stress-strain curves were calculated from the measured force and the true cross-sectional area of the deformed specimen by the following equations:

퐹푡 퐹푡 휎푇 = = 2 (4.1) 푊푡퐷푡 푊푡 [( )퐷0] 푊표

2 퐿푡−퐿0 푊표 휀푇 = = ( ) − 1 (4.2) 퐿0 푊푡

2 퐿푡 푊0 휀퐻 = 퐿푛 ( ) = 퐿푛 ( ) (4.3) 퐿0 푊푡

where 휎푇, 휀푇, and 휀퐻 are the true stress, true strain, and Hencky strain, respectively. F, D,

W, and L are the force, thickness, width, and length, respectively. Subscript o indicates initial values whereas subscript t indicates real time values. All the equations above were determined by measuring the real-time width of the specimen and assuming an incompressibility and transverse isotropy:

퐷0푊0퐿0 = 퐷푡푊푡퐿푡 (4.4)

푊 퐷 푡 = 푡 (4.5) 푊0 퐷0

A visible wavelength light source was used to measure the optical retardation () with resolution of ± 2nm through the polymer film. The in-plane birefringence (n12) was calculated from the following equation:

훤 훤 ∆푛12 = = 푊푡퐷0 (4.6) 퐷푡 ( ) 푊0

The crystal structures were characterized by wide-angle X-ray diffraction

(WAXD). WAXD measurements were performed using a Bruker AXS D8 Goniometer.

The instruments used CuKα radiation (λ = 0.1542 nm). WAXD diffractograms were used to measure any changes in the crystal structure of PI(BTDA-DAH) upon stretching.

A TA Instruments differential scanning calorimeter (DSC Q-200) was used to measure the glass transition, cold crystallization range, melting point (Tm) and the heat of fusion (Hf) of PI(BTDA-DAH) films. Before the DSC measurements, temperature calibration was performed using an Indium standard (Tm = 156.6C). A heating rate of

10ºC/min was employed. The melting point was defined as the peak of the melting endotherm, with the largest peak denoted by Tm. The mass fraction crystallinity of the polymer (m) was calculated from the following ratio:

∆퐻푓−∆퐻푐 휙푚 = (4.7) ∆퐻푃퐼

where Hf is the area under the DSC melting endotherm, Hc is the area under the

DSC cold crystallization exotherm, and HPI is the enthalpy of fusion of a pure Polyimide crystal which is 144.14 J/g. [123]

4.3 Results and Discussion

The thermal transitions for PI(BTDA-DAH) are shown by the DSC curve in Figure

4.2. This polymer is considered to be a slow-crystallization polymer as it exhibits three transition temperatures--glass, cold crystallization, and melting--similar to PLA [27,124],

PET, [125] PPS, [126] and PEN. [28] The cast films exhibited a glass transition of 52.2°C, cold crystallization of 159.2°C, and melting temperature of 247.73°C. The areas under the cold crystallization and melting peak were essentially identical; therefore, the films were amorphous after solution casting. Usually slow-crystallizing polymers are processed in their rubbery state between the glass transition and cold crystallization transitions, where they exhibit very low thermal crystallization rates due to high rubbery viscosity that suppress polymer chain diffusion. [27-29] Therefore, the material generally exhibits strain hardening during stretching, which is essential for obtaining uniform films. [26]

0.0

Tcc=159.2°C

ΔHc=18.01 J/g -0.5 Tg=52.2°C

-1.0

stretching -1.5 widows Tm=247.7°C ΔHf=20.96 J/g Heatflow (w/g)(endo down) Φ=1.79%

-2.0 50 100 150 200 250

o Temperature ( C)

FIGURE 4.2 DSC THERMOGRAMS FOR PI (BTDA-DAH) POLYMER OBTAINED FROM SOLUTION-CAST AND DRIED FILMS. THE DASHED LINES INDICATE THE IDEAL STRETCHING WINDOW IN WHICH THE THERMAL CRYSTALLIZATION RATES ARE VERY LOW. THE SYMBOLS INDICATE THE FOLLOWING: TG – GLASS TRANSITION TEMPERATURE, TCC – COLD CRYSTALLIZATION TEMPERATURE, TM – MELTING POINT., ΔHC – HEAT OF COLD CRYSTALLIZATION, AND ΔHF – HEAT OF FUSION.

4.3.1 Mechanical Behaviors

Figure 4.3 and 4.4 show the true stress-true (Hencky) strain curves of PI(BTDA-

DAH) films during uniaxial deformation at different temperatures and stretch rates, respectively. In the data illustrated in Figure 4.3, the polymer films were heated to varying temperatures between 45°C to 105°C and isothermally maintained at those temperatures for 5 minutes. The films were then stretched at a rate of 0.0024sec-1. All the specimens showed an elastic deformation followed by yielding and strain hardening. Young’s modulus

(E) and the yield strength (σy) naturally decreased upon increasing the deformation temperature. At deformation temperatures up to 80°C, an elastic deformation was observed, followed by yielding and then strain hardening. The yielding stress decreased upon increasing the deformation temperature. At temperatures of 80°C and higher, only plastic deformation was observed, followed by strain hardening. In Figure 4.4, all the polymer films were isothermally stretched at varying rates between 0.0024sec-1 to 0.24sec-1 at

105°C. At higher strain rates, the films showed high plateau stresses. Furthermore, the critical stress at which the strain hardening started decreased with increasing strain rates.

100 45°C 52°C 80

60°C 60

Pa) 65°C

T 40 80°C 105°C

0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

FIGURE 4.3 TRUE STRESS-HENCKY STRAIN PLOTS FOR PI(BTDA-DAH) FILMS AT VARIOUS TEMPERATURES. ALL THE FILMS WERE HEATED TO STRETCHING TEMPERATURES AND ISOTHERMALLY MAINTAINED AT THAT TEMPERATURE FOR 5 MINUTES BEFORE STRETCHING AT A CONSTANT RATE OF 0.0024SEC-1.

0.0048 sec-1

30 0.024 sec-1

0.048 sec-1

-1 20 0.24 sec

Pa)

0.0024 sec-1 0 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

FIGURE 4.4. TRUE STRESS-HENCKY STRAIN PLOTS FOR PI(BTDA-DAH) FILMS AT VARIOUS RATES. ALL THE FILMS WERE HEATED TO 105°C AND ISOTHERMALLY MAINTAINED AT THAT TEMPERATURE FOR 5 MINUTES BEFORE STRETCHING.

4.3.2 Mechano-Optical Behavior

Figure 4.5 shows true stress birefringence behavior of solution cast PI(BTDA-

DAH) specimen films during uniaxial stretching at a constant rate of 0.0024sec-1 and varying temperatures. As has been shown elsewhere, [27] birefringence behavior during uniaxial deformation can follow I-II-IIIc behavior or I-IIIa behavior. In I-II-IIIc behavior, the birefringence slowly increases at a low level of deformation (regime I). In this regime, the birefringence is linear and therefore the stress optical rule Δn=Cσ is valid, where Δn is the birefringence, C is the stress optical coefficient, and σ is stress. After a critical point,

which is the end of regime I, a sharp increase in birefringence occurs at a very low level of stress, which is called regime II. This deviation (i.e. regime II) from linearity is due to the strain-induced crystallization phenomenon. [27-29] Regime III is characterized by a slow increase in birefringence (negative deviation) until a plateau is reached. This negative deviation (i.e. regime III) is a result of polymer chains reaching the limit of their extensibility. [27-29] In I-IIIa behavior, the birefringence linearly increases with stress

(regime I); this is followed by a negative deviation from linearity, which slowly increases in birefringence until a plateau is reached (regime IIIa). In this case the strain-induced crystallization does not occur. This would indicate that at the end of regime IIIa the polymer chains are organized in a structure with a nematic order. [27,29] Both of these behaviors can be accompanied by a glassy component/behavior before regime I. [29] The effect of temperature and stretching rate on stress-birefringence behaviors are shown in Figure 4.5

(a) and 4.6 respectively. At temperatures lower than 80°C, a glassy component was observed, as shown in Figure 4.5(a) and (b). Therefore, stress increased with low changes in birefringence. This component decreased as the stretching temperature increased, until it completely disappeared at 80°C and above. For all the films in which a glassy component existed, a I-IIIa behavior was observed. However, for PI(BTDA-DAH) films in which a glassy component did not exist, a I-II-IIIc behavior was observed. For those PI(BTDA-

DAH) films, the stress optical constant was 17.8 GPa-1. In cases where I-II-IIIc behavior existed, the critical stress point at which birefringence deviated from linearity decreased with temperature while the critical birefringence point increased, as shown in Figure 4.7

Furthermore, this positive deviation slope was larger with higher temperatures. At higher temperatures the polymer exhibited more rubber-like behavior. Therefore, it was easier for

the polymer chains to orient parallel to each other and relax, [27] which is necessary for strain-induced crystallization. For example, at temperatures below 80°C, the PI(BTDA-

DAH) films were in a glassy state or had glassy components; for this reason, the stress relaxation phenomenon did not occur during stretching. While increasing the temperature, the polymer PI(BTDA-DAH) films were in a rubbery state and all the glassy components were already melted, which allowed stress relaxation to compete the orientation. Therefore, the critical stress decreased with an increase in temperature, as can be seen in Figure 4.7.

Mulligan et. al. observed that I-IIIa behavior appeared at high stretching rates even in a rubbery state when the temperature was high. [27] The same observations were found for other slow crystallization polymers. [127] Figure 2.17 The dependency of true stress birefringence behavior on stretching rates for solution cast PI(BTDA-DAH) specimen films during uniaxial stretching at 105°C can be observed in Figure 4.6. Although the maximum stretching rate was 0.24 sec-1 (500 mm/min) in this study, the I-IIIa regime was not observed. This indicates that at a stretching temperature of 105°C and stretching rates of at least up to 0.24 sec-1, the polymer chains still have time to relax.

0.20 a 80°C 0.18 105°C 65°C 60°C 0.16 45°C

0.14 52°C 0.12

n 0.10

0.08

0.06

0.04

0.02 SOC=17.8 GPa-1 0.00 0 20 40 100 200 300 400 (MPa) T

FIGURE 4.5 TRUE STRESS-BIREFRINGENCE BEHAVIOR OF SOLUTION CAST POLYIMIDE(BTDA-DAH) FILMS DURING UNIAXIAL DEFORMATION AT A CONSTANT RATE OF 0.0024SEC-1 FOR A) VARYING TEMPERATURES AND B) DIFFERENT REGIMES OF STRESS BIREFRINGENCE BEHAVIOR. ALL THE FILMS WERE HEATED TO STRETCHING TEMPERATURES AND ISOTHERMALLY MAINTAINED AT THAT TEMPERATURE FOR 5 MINUTES BEFORE STRETCHING.

FIGURE 4.6 TRUE STRESS-BIREFRINGENCE BEHAVIOR OF SOLUTION CAST PI(BTDA- DAH) DURING UNIAXIAL DEFORMATION AT A CONSTANT TEMPERATURE OF 105°C AND VARYING RATES. ALL THE FILMS WERE HEATED TO 105°C AND ISOTHERMALLY MAINTAINED AT THAT TEMPERATURE FOR 5 MINUTES BEFORE STRETCHING.

6 0.07

0.06

5 0.05

Pa)

0.04

T 4 0.03

Critical 0.02 3

Criticalbirefringence

0.01

2 0.00 50 60 70 80 90 100 110

o Temperature ( C)

FIGURE 4.7 CRITICAL VALUES OF TRUE STRESS AND BIREFRINGENCE AT THE POINT OF DEVIATION FROM LINEARITY OF STRESS OPTICAL RULE OF SOLUTION CAST PI(BTDA- DAH) FILMS DURING UNIAXIAL DEFORMATION AT A CONSTANT RATE OF 0.0024SEC-1 AND VARYING TEMPERATURES.

The photoelastic constant (Cʹo) and regime I's slopes are plotted as a function of temperature in Figure 4.8. The regime I slopes tend to decrease with an increase in temperature, while the photoelastic constant increases with the temperature. The regime I slope overlaps and merges with the photoelastic constant at 80°C. This is the critical temperature at which the glassy component is fully melted and the stress optical rule is valid.

)

-1

30 slope regime

stress optical constant 20

photoelastic constant 10

Photoelastic& Stress optical costant (GPa 50 60 70 80 90 100 110

o Temperature ( C)

FIGURE 4.8 PHOTOELASTIC AND REGIME I SLOPE AS A FUNCTION OF TEMPERATURE OF SOLUTION CASTING PI(BTDA-DAH) FILMS DURING UNIAXIAL DEFORMATION AT STRESS RATE 0.0024SEC-1.

4.3.3 Strain Optical Behavior

Figure 4.9 and 4.10 show the true strain optical behavior of solution cast PI(BTDA-

DAH) specimen films during uniaxial stretching at varying temperatures and rates

respectively. An almost linear dependency between strain and birefringence was observed at temperatures below 80°C. At this critical temperature and above, a change in the shape of the curve was observed. The change in the shape related to higher degrees of relaxation in the polymer specimen films at higher temperatures. [27] This critical point matches a transition range that was observed by Boyer et al. and was called a liquid-liquid transition,

Tll. [128,129] Boyer suggested that Tll is a kinetic phenomenon with a thermodynamic aspect, and that it occurs at ~1.2-1.25Tg (K). Above this temperature, a relaxation of amorphous chains occurs. In short, below the Tll segments of the polymer chains behave like anchor points (chain entanglements) which suppress the relaxation of polymer chains.

Above this temperature, this "segmental melting" [130] occurs and the material behaves as a true liquid. Tll transition is still controversial; [131] however, more researchers have recently started to accept this transition temperature. [130-133] For example, Kisluk et al. described a similar transition temperature at 1.15-1.35Tg; however, they defined this phenomenon as dynamic. [132] In our study, we assume this temperature transition is real and that it occurs at 1.08Tg (K). Although this temperature is lower than that suggested by

Boyer as well as Kisluk et al., it is close, and even higher than those found for Polyethylene oxide. [133] Similar dependency of strain birefringence behavior on strain rates was noticed. At higher strain rates an almost linear dependence between strain and birefringence was observed for temperatures higher than Tll.

FIGURE 4.9 HENCKY STRAIN-BIREFRINGENCE BEHAVIOR OF SOLUTION CAST PI(BTDA- DAH) FILMS DURING UNIAXIAL DEFORMATION AT A CONSTANT RATE OF 0.0024SEC-1 AND FOR VARYING TEMPERATURES. ALL THE FILMS WERE HEATED TO STRETCHING TEMPERATURE AND ISOTHERMALLY MAINTAINED AT THAT TEMPERATURE FOR 5 MINUTES BEFORE STRETCHING.

FIGURE 4.10 HENCKY STRAIN-BIREFRINGENCE BEHAVIOR OF SOLUTION CAST PI(BTDA-DAH) FILMS DURING UNIAXIAL DEFORMATION AT A CONSTANT TEMPERATURE OF 105°C AND FOR VARYING STRETCH RATES. ALL THE FILMS WERE HEATED TO STRETCHING TEMPERATURE AND ISOTHERMALLY MAINTAINED AT THAT TEMPERATURE FOR 5 MINUTES BEFORE STRETCHING.

4.3.4 Structural Studies: Effect of Deformation

Off-line experiments (x-ray and DSC) have been performed on a series of

PI(BTDA-DAH) specimen films that were stretched at 105°C and 0.0024sec-1 to different deformation levels. These experiments clarified the structural mechanisms responsible for the different mechano-optical behaviors observed during uniaxial stretching. Figure 4.11 shows the mechano-optical behavior of series of PI(BTDA-DAH) specimen films, together with WAXD patterns and levels of crystallinity determined by DCS. As was discussed above, the PI(BTDA-DAH) specimen films showed I-II-IIIc behavior at these processing conditions, temperatures and rates. In regime I, the material remained amorphous, as indicated by low crystallinity levels (below 3%) and by a broad halo in the WAXD pattern.

In an early stage in regime II the level of crystallinity slightly increased. The WAXD pattern showed equatorial peaks, which indicated a level of chain orientation. Furthermore, along the meridian direction substantial streak diffractions were observed. In the end of regime II, the crystallinity level increased drastically to ~20%. The WAXD pattern showed sharp equatorial peaks which indicates of high level of chain orientation. Moreover, 5 substantial streak diffractions were observed along the meridian direction. This pattern was observed in the past in different polymers. [134-137] This phenomenon can be explained by a periodic disordered conformation of polymer chains in the crystal structure and an irregularity in the packing of the chains in the lateral direction with respect to the fiber direction. [134,137] Almost no change of the WAXD pattern and the crystallinity level were noticed after the critical point between regime II and IIIc. The WAXD pattern still showed very sharp equatorial peaks as well as substantial 5 streak diffractions along the

meridian direction. Furthermore, the absence of off-equatorial diffraction spots revealed that a three-dimensional crystalline order was absent. This suggests that the three- dimensional crystalline order does not exist even at regime IIIc. The absence of the three dimensional crystalline order in regime IIIc is unique when compared to other slow crystallization polymers. [27-29] The reason for this absence might be the disorder in the crystal structure and the irregularity in the packing of the polymer chains.

FIGURE 4.11 MECHANO-OPTICAL BEHAVIOR OF SOLUTION CAST PI(BTDA-DAH) FILMS DURING UNIAXIAL DEFORMATION AT A CONSTANT TEMPERATURE OF 105°C, STRAIN RATE OF 0.0024SEC-1, AND DIFFERENT STRETCH RATIOS WITH CORRESPONDENT WAXD PATTERNS AND % OF CRYSTALLINITY.

100

Figure 4.12 shows DSC thermogram curves of PI(BTDA-DAH) polymer films that were stretched uniaxially at different stretch ratios at a constant temperature of 105°C and a strain rate of 0.0024sec-1. Figure 4.13, on the other hand, shows DSC thermogram curves of PI(BTDA-DAH) polymer films that were uniaxially stretched at different stretching temperatures at a constant strain rate of 0.0024sec-1. The glass transition temperature shifted to a higher temperature when the stretch ratios and temperatures were increased.

This is due to the combination of the evaporation of the solvent trapped in the polymer film with the elevated arrangement of the polymer chains. The polymer chains were packed during the uniaxial deformation, which decreased the chain mobility. Increasing the stretch ratios made the cold crystallization temperature shift to lower temperatures. This change was observed until the cold crystallization peak combined with the glass transition and disappeared. This shift and the disappearance of the cold crystallization peak indicated decreases in entropy due to a higher arrangement of the polymer chains in the amorphous region. These data and the values of ∆H and crystallinity are summarized in Table 4.1 and

4.2. The cold crystallization peak of PI(BTDA-DAH) polymer films which were stretched at varying temperatures shifted to lower temperatures when the stretching temperatures were increased. The cold crystallization disappeared when the specimens were stretched above the critical point, as can be seen in Figure 4.13and Table 4.2

101

FIGURE 4.12 DSC THERMOGRAM CURVES FOR PI(BTDA-DAH) POLYMER FILMS. SAMPLES STRETCHED UNIAXIALLY AT DIFFERENT STRETCH RATIOS, A CONSTANT TEMPERATURE OF 105°C, AND A STRAIN RATE OF 0.0024SEC-1.

FIGURE 4.13 DSC THERMOGRAM CURVES FOR PI(BTDA-DAH) POLYMER FILMS. SAMPLES STRETCHED UNIAXIALLY AT DIFFERENT STRETCH TEMPERATURES AT A CONSTANT STRAIN RATE OF 0.0024SEC-1.

102

TABLE 4.1 THERMAL PROPERTIES OF PI(BTDA-DAH) POLYMER FILMS AFTER UNIAXIAL STRETCHING AT DIFFERENT STRETCHING RATIOS AT A CONSTANT TEMPERATURE OF 105°C AND STRAIN RATE OF 0.0024SEC-1.

Glass Cold Melting Heat of fusion Crystallinity Stretching transition crystallization transition ratio Tcc' Tcc Tm' Tm Ho Hf  Tg (C) (C) (C) (C) (C) As cast 53.42 123.27 159.2 199.92 247.7 18.01 20.60 1.79

0.8 81 140.1 151.93 205 254.65 16.07 20.73 3.23

1.3 78.48 104.97 121.98 212.80 253.45 4.752 24.20 13.49

1.8 78.61 - - 211.44 255.84 - 25.53 17.71

2.3 89.58 - - 211.88 255.66 - 29.81 20.68

2.47 92.61 - - 201.23 256.03 - 27.53 19.1

2.6 87.99 - - 202.11 251.75 - 31.08 21.56

TABLE 4.2 THERMAL PROPERTIES OF PI(BTDA-DAH) POLYMER FILMS AFTER UNIAXIAL STRETCHING AT VARYING TEMPERATURES AND A STRAIN RATE OF 0.0024SEC- 1. Glass Cold Melting Heat of Crystallinity Stretching transition crystallization transition fusion temperature Tcc' Tcc Tm'  Ho Hf  (C) Tg (C) (C) (C) (C) As cast 53.42 123.27 159.2 199.92 247.7 18.01 20.60 1.79

45 52.87 75.69 90.17 201.71 3.23 5.357 24.06 12.97

52 58.82 76.02 91.98 203.19 13.49 7.49 24.05 11.48

60 58.27 78.47 91.32 207.2 17.71 3.826 25.67 15.15

65 59.16 78.47 92.07 207.52 20.68 3.458 23.57 13.95

80 54.7 - - 208.49 19.1 - 34.32 23.81

105 87.99 - - 209.46 21.56 - 31.08 21.56

103

4.3.5 Structural Evolution

A schematic representation of the structural evolution of PI(BTDA-DAH) from an amorphous state to a final crystalline state during uniaxial stretching at above a Tll temperature with low stretching rates based on the above results and discussions is provided in Figure 4.14. At the beginning of deformation, the polymer is in an amorphous state. The polymer chains are unoriented and this order remains until the critical point between regime I and regime II. In regime II the polymer chains start to orient and strain- induced crystallization is promoted. Although the crystallization is promoted, which in turn tightens the network structure and results in a fast increase in birefringence, an irregular order in the network structure exists, as can be seen by streaks in the meridian direction in the WAXD diffractions. At regime IIIc the polymer chains approach their extension limitation; there is, therefore, negative deviation in birefringence (the birefringence increases very slowly) while the stress continues to increase. The irregular order in the polymer network structure can also be seen in regime IIIc.

104

Crystallinity (%) Crystallinity

FIGURE 4.14 SCHEMATIC REPRESENTATION OF THE STRUCTURAL EVOLUTION OF PI(BTDA-DAH) DURING UNIAXIAL STRETCHING.

105

4.4 Conclusions

The effect of processing conditions, temperatures, and stretching rates on the mechano-optical behavior of PI(BTDA-DAH) at glassy and rubbery states were studied.

At stretching temperatures above Tll and low stretching rates, three regimes of stress optical behavior were observed. The first regime was linear and followed the stress optical rule; the stress optical constant was found to be 17.8 GPa-1. In this regime the polymer structure remained amorphous. In regime II, a rapid increase in birefringence with a modest increase in stress was observed; it was caused by strain-induced crystallization. Furthermore, irregularity in the network structure order was observed in WAXD patterns. In regime IIIc the polymer chains reached their extension limitation; therefore, the birefringence leveled off with a high increase in stress. The PI(BTDA-DAH) did not develop a three dimensional crystalline order during the stretching. At stretching temperatures below Tll the stress optical behavior follows regimes I-IIIa.

106

CHAPTER V

ROLE OF RELAXATION ON STRAIN INDUCED CRYSTALLIZATION OF

UNIAXIALLY STRETCHED PI(BTDA-HDA) FILMS: REAL-TIME INFRARED-

MECHANO-OPTICAL BEHAVIOR AND ELECTRICAL STUDY

This chapter focuses on the real-time rheo--mechano-optical behavior of PI(BTDA-

HAD) during uniaxial deformation and relaxation. The rheo-mechano-optical behavior of the polymer is studied at 100°C while at different strain levels of deformation during uniaxial stretching followed by relaxation. At this temperature the polymer is in its rubbery state. This study is performed using an instrumented system that combines real-time spectral birefringence and URS-FTIR measurements during uniaxial deformation. The true stress, true strain, birefringence, and orientation function of the PI(BTDA-DAH) are measured by this instrument. Three regimes of stress optical behavior as well as stress orientation function behavior are revealed. First, at the early stage of deformation (regime

I) the stress optical rule is observed; birefringence and orientation function linearly increase with stress. Second, in regime II, a positive deviation from linearity takes place and the birefringence and orientation function rapidly increase while the stress slowly increases.

Third, in regime IIIc, a negative deviation of birefringence and orientation function from linearly are observed which reach a plateau while stress rapidly increases. Both

107

birefringence and orientation function behavior during relaxation depend on the level of deformation. At deformation in regime I, birefringence and orientation function linearly decrease with stress. At an early stage of deformation in regime II, they slowly decrease.

This behavior changes with the deformation level in regime II. At an intermediate stage of the deformation in regime II, the birefringence and orientation function do not change while the stress rapidly decreases; the birefringence and orientation function then slowly start to increase while the stress continues to decrease. At the high stage of the deformation in regime II, the birefringence and orientation function slowly increase while the stress rapidly decreases, and then they rapidly increase while the stress slowly decreases. At the high stage of the deformation in regime III, the birefringence and orientation function slowly increase while the stress rapidly decreases, and then they slowly increase while the stress continues to decrease. Relaxation time also affects the birefringence and orientation function behaviors. With enough relaxation time, they increase at all levels of deformation in regime II. According to off-line characterization techniques (DSC and WAXD), an increase in the crystallinity was observed during relaxation in regime II. This implies stress-induced crystallization associated with regime II. During uniaxial stretching, no significant change in the crystallinity is observed in regime III; however, the crystallinity continues to increase during relaxation.

108

5.1 Introduction

The study of energy storage systems has received much attention lately.

Electromagnetic (EM) railguns [78.138] and Hybrid cars [110,139] are just some of the products of this new focus. The different storage systems are characterized by the amount of energy that they can store and the system's rate of energy transfer. [2] One of the systems that attracts researchers is capacitors because of their relatively short charge time and high energy release rates. [2,3]

Today's metalized BOPP is the most popular polymer for capacitor application.

Although it is the most popular polymer, it suffers from a low dielectric constant of ~2.2,

[79-81] a low-energy density of 5 at breakdown (which occurs at ~720V/μm for films ~10

μm thick), [80] and a reduction of breakdown strength above 85°C. [113] Therefore, efforts have increased to develop novel polymers with high-temperature stability, a high dielectric constant, high breakdown strength, and low dielectric loss to replace BOPP. [119, 140-

142]

Due to its a high dielectric constant, a high breakdown strength, and a high range of temperature stability, [119] PI(BTDA-DAH) is one of the new polymers which may be suitable to replace BOPP for capacitor applications.

PI(BTDA-DAH) is a slow crystallization polymer. [74] It exhibits three transition temperatures: glass, cold crystallization, and melting. Slow-crystallization thin polymeric films in the micrometer range are usually prepared by a melt casting line (a tenter-frame biaxial stretching machine). In this processing method, the polymers which are melted in

109

the extruder flow through a slit-shaped die and quench below their glass transition temperature. Therefore, the thermal crystallization phenomenon is suppressed, which leads to amorphous polymers. [27] Then, the polymers are heated between the glass transition temperature and the cold crystallization temperature. They are then stretched in the MD, and afterward cooled. In the temperature stretching range in which they are heated, the thermal crystallization rates are very low due to the high viscosity that suppress polymer chain diffusion. [27-29] Therefore, the polymers generally exhibit the strain hardening that is essential for obtaining uniform films. [143] Furthermore, the polymer films undergo a relaxation phenomenon before cooling. The polymers are heated once again between the glass transition temperature and the cold crystallization temperature before being stretched in the TD; this is followed by heat setting and cooling. Because relaxation and orientation happen simultaneously, they have a simultaneous yet opposing affect on the conformation of polymer chains; [28] therefore, the polymers are subject to complex deformation behavior, which impacts the final state of polymer chain orientation, the amount of crystallinity, and final film performance.

The effects of temperature, stretching rates, and stretching ratios on the deformation behavior of PI(BTDA-DAH) during uniaxial stretching were investigated. [74] However, the effect of relaxation on stretched, pre-oriented PI(BTDA-DAH) films has not been investigated. In this study, on-line measurements of true stress, true (Hencky) strain, birefringence, and FTIR dichroism are performed alongside off-line WAXD, DSC, and

IMASS Time Domain Dielectric Spectrometer measurements to investigate the effect of relaxation on PI(BTDA-DAH) physical structure as well as PI(BTDA-DAH) mechanical, optical, and dielectrical properties.

110

5.2 Experimental Details

5.2.1 Materials

1,000 ml of M-Cresol, 110.2g (0.342 mole) of 3,3′,4,4′-

Benzophenonetetracarboxylic dianhydride (BTDA) are added to a clean a dried 3,000L three-neck flask under nitrogen with magnetic stirring. After stirring thoroughly for 30 minutes, 39.74g of 1,6 Hexane-diamine (HDA) is added to the stirring solution under nitrogen gas. The reaction is carried out at 30°C for 24 hours followed by imidization at

165-175°C for 6 hours. The solution is precipitated into 10,000 ml of methanol under vigorous mechanical stirring and forms continuous fibers. After soaking for 6 hours the fibers are stirred several times in 2,000 ml portions of methanol to further remove M-

Cresol. The fibers are dried at room temperature under a constant stream of air for 48 hours before being vacuum dried for 24 hours at room temperature. Light yellow fibers are obtained in a 92% yield. The process was repeated several times to obtain roughly one kilogram of total product. A schematic of this synthesis is showed in Figure 5.1

111

M-Cresol 30°C 24 hours

M-Cresol 165°C-175°C 6 hours

FIGURE 5.1 SYNTHESIS OF PI (BTDA-DAH) POLYMER.

5.2.2 Sample Preparation

A 13 wt.% solution of PI(BTDA-DAH) in M-Cresol is made. The solution is filtrated through a 0.45 µm PTFE filter. Unoriented films are prepared by a casting procedure. The solution is cast on a PET substrate (Mylar) in a roll to roll casting line by using a 6'' wide casting doctor blade. The films are dried for 48 hours at room temperature and then at 60°C for another 12 hours. In order to peel the PI(BTDA-DAH) films from the

PET substrate, the films are kept in DI water for 3½ hours. A final thickness of 15 µm is obtained. Dumbbell-shaped and rectangular specimens are cut for further characterization.

112

5.2.3 Characterization

PI(BTDA-DAH) specimens in their rubbery states are uniaxially stretched at a rate of 10 mm/min at 100°C. A real-time mechano-optical measurement platform equipped with an URS polarized FTIR spectrometer is used to study the deformation behavior of

PI(BTDA-DAH) during stretching and relaxation. Details of the measurement system are discussed elsewhere. [89] In brief, a specimen is clamped to an uniaxial stretching machine enclosed by a heating chamber. During the stretching experiment, both of the rods which hold the specimen move simultaneously at the same speed; therefore, the middle section of the polymer specimen stays stationary. By assuming incompressibility ( 퐷0푊0퐿0 =

퐷푡푊푡퐿푡) and transverse isotropy (푊푡/푊0 = 퐷푡/퐷0), as well as by measuring the real-time width of the specimen, the stress-strain curves are calculated by the following equations:

푭풕 푭풕 흈푻 = = ퟐ (5.1) 푾풕푫풕 푾풕 [( )푫ퟎ] 푾풐

ퟐ 푳풕 푾ퟎ 휺푯 = 푳풏 ( ) = 푳풏 ( ) (5.2) 푳ퟎ 푾풕

where 휎푇is the true stress and 휀퐻 is the Hencky strain. F, D, W, and L are force, thickness, width, and length respectively. The notations o and i indicate initial value and real time value respectively.

A visible wavelength light source is used to measure the optical retardation () with resolution of ± 2nm through the polymer film. The in-plane birefringence (n12) is calculated using the following equation:

113

휞 휞 ∆풏ퟏퟐ = = 푾 푫 (5.3) 푫풕 ( 풕 ퟎ) 푾ퟎ where the retardation wavelength that is used is 546nm.

URS-polarized FTIR is used to measure the infrared (IR) absorbance at a rate of

100 spectra per second. This high rate scan is possible by using a disk mirror interferometer instead of a Michelson/standard IR interferometer. [89] Inside the mirror interferometer, the IR beam is polarized into two beams by a beam splitter. By using the absorbance of the polarized IR light, the dichroic ratio (퐷) and Herman’s orientation function (fIR) orientation function can be found by:

퐀 푫 = ∥ (5.4) 퐀⊥

(푫−ퟏ)(푫ퟎ+ퟐ) 풇푰푹 = (5.5) (푫+ퟐ)(푫ퟎ−ퟏ)

where D0 is the dichroic ratio of an ideally uniaxial oriented polymer, which is defined as:

ퟐ 푫ퟎ = ퟐ 퐜퐨퐭 흍 (5.6) where ψ is the transition moment angle of a specific side-group vibration with respect to the polymer chain axis.

WAXD measurements are performed using a Bruker AXS D8 Goniometer. The instruments use CuKα radiation (λ = 0.1542 nm). WAXD diffractograms are used to measure any changes in the crystal structure of the PI(BTDA-DAH) upon stretching and relaxation.

A TA Instruments differential scanning calorimeter (DSC Q-200) is used to measure the glass transition, cold crystallization range, melting point (Tm) and heat of

114

fusion (Hf) of PI(BTDA-DAH) films. Before the DSC measurements, temperature calibration is performed using an Indium standard (Tm = 156.6C). A heating rate of

10ºC/min is employed. The melting point is defined as the peak of the melting endotherm, with the largest peak denoted by Tm. The mass fraction crystallinity of the polymer (m) is calculated from the following ratio:

∆푯풇−∆푯풄 흓풎 = (5.7) ∆푯푷푰 where Hf is the area under the DSC melting endotherm, Hc is the area under the DSC cold crystallization exotherm, and HPI is the enthalpy of fusion of a pure PI crystal. The

HPI was calculated by using the Flory melting point depression theory for solvent polymer systems: [144-147]

ퟏ ퟏ − 푻 푻 풐 푹푽 푩푽 흑 풎 풎 = 풖 (ퟏ − ퟏ ퟏ) (5.8) 흑ퟏ ∆푯풇푽ퟏ 푹푻풎

o where Tm is the melting point of the polymer-solvent system, Tm is the equilibrium melting point of pure PI (BTDA-DAH) polymer, Vu is the molar volume of the polymer, V1 is the molar volume of the solvent, ∆Hf is the heat of fusion of 100% crystalline, R is the gas constant, B is the interaction energy density of the solvent-solute pair, and ϑ1 is the volume fraction of diluent.

An IMASS Time Domain Dielectric Spectrometer with a novocontrol GmbH concept 40 broadband dielectric spectrometer is used to study the dielectric constant and loss. 10 V AC at varying frequencies of 10-3 to 104 and temperatures of R.T, 50°C, 75°C,

100°C, 125°C, and 150°C are applied to measure the dielectric constant and loss.

115

Polarization measurements are conducted with a modifed Sawyer-Tower circuit, employing a Trek Model 10/40 10 kV high voltage amplifer and an OPA541 operational amplifer based current to voltage converter. The samples are sputtered with Au/Pd

80/20wt% electrodes of 0.07 cm2.

5.3 Results and Discussion

The thermal transitions for PI (BTDA-DAH) are shown by the DSC curve in Figure

5.2. Three transition temperatures exist: glass transition temperature, cold crystallization temperature and melting temperature. The areas under the cold crystallization and melting peak are essentially identical; therefore, the films are considered to be amorphous.

116

Tcc=171.4°C Tg=80.5°C stretching ΔHc=19.5 J/g temperature

Tm=243.3°C

ΔHf=19.9 J/g =~0.3%

stretching windows

FIGURE 5.2 DSC THERMOGRAM CURVES FOR PI (BTDA-DAH) POLYMER OBTAINED FROM UNDEFORMED FILM. THE DASH MARKS INDICATE THE STRETCHING WINDOW IN WHICH THE THERMAL CRYSTALLIZATION RATES ARE SLOW. THE SYMBOLS INDICATE THE FOLLOWING: TG – GLASS TRANSITION TEMPERATURE, TCC – COLD CRYSTALLIZATION TEMPERATURE, TM – MELTING POINT, ΔHC – HEAT OF FUSION OF COLD CRYSTALLIZATION PEAK, AND ΔHF – HEAT OF FUSION,-AMOUNT OF CRYSTALLINITY.

5.4.1 Depression of melting point

The heat of fusion of 100% crystallinity of PI(BTDA-DAH) is calculated by using melting temperature depression theory. [147] Melting temperature depression is a statistical method which calculates the heat of fusion of 100% crystallinity of PET, PVA,

ퟏ ퟏ − 풐 흑 PE, and others. [144-146] By plotting 푻풎 푻풎 vs ퟏ according to equation 5.8, the heat of 흑ퟏ 푻풎

117

fusion of 100% crystallinity of PI (BTDA-DAH) is obtained from the intercept. This intercept was found to be 144.14 (J/g), as can be seen in Figure 5.3.

1.0 퐽 ∆퐻 = 144.14 ( ) 푓−푃퐼 푔 0.8

]x10 0.6

- 0.4

[ 0.2

0.0 0.0 0.1 0.2 0.3 0.4

3 (흑 / T )x10 ퟏ T m

ퟏ ퟏ − 푻 푻 풐 흑 FIGURE 5.3 PLOT OF 풎 풎 VS ퟏ FOR POLYIMIDE (BTDA-DAH) WITH M-CRESOL. 흑ퟏ 푻풎

5.3.2 Mechanical Behavior

Figure 5. 4 show true stress-true (Hencky) strain curves of PI(BTDA-DAH) films when they have undergone uniaxial deformation followed by 10 minutes of relaxation. The

PI(BTDA-DAH) specimens are stretched at a rate of 10 (mm/min) to different stretching ratios. Prior to stretching, the PI(BTDA-DAH) specimens are heated to 100°C and kept at this temperature for 20 minutes. In this study, the relaxation temperature is kept the same as stretching temperature.

118

During stretching, the polymer films exhibit an elastic deformation followed by yielding and then strain hardening. As expected, the stress drops during relaxation.

Although the PI(BTDA-DAH) specimen films do not retract during relaxation, a change in the Hencky strain is observed. This observation could not have been seen in engineering mechanical behavior, as can be seen in Figure 5. 4. The change in the Hencky strain can be explained as the result of the spontaneous deformation which occurs during relaxation.

Similar observations are reported by Martins at el. for PEN films. [148]

2.5

2 1 1.25 1.5 0.1 0.5

FIGURE 5. 4 TRUE STRESS-HENCKY STRAIN PLOTS FOR PI(BTDA-DAH) FILMS DURING THE STRETCHING AND RELAXATION PROCESS. SAMPLES WERE STRETCHED BETWEEN 10% TO 250% BEYOND THEIR ORIGINAL LENGTH AT 100°C AND RELAXED FOR 10 MINUTES. IN THE INSET, ENGINEERING MECHANICAL BEHAVIOR IS PRESENTED; THE SPONTANEOUS DEFORMATION CANNOT BE OBSERVED.

119

The changes in the Hencky strain’s behavior during relaxation are impacted by the level of strain deformation during the uniaxial stretching. In order to understand these changes, the Hencky strain is plotted as a function of relaxation time in Figure 5.5. The strain values of the PI(BTDA-DAH) specimen films are normalized to their initial strain value at relaxation time zero. This makes possible a comparison between relaxing

PI(BTDA-DAH) specimen films that are stretched to different strain levels. Two tendencies in the change of deformation are observed: a slow increase in the strain which occurred at low and high stretching levels of deformation, and a fast increase in the strain through the beginning of relaxation which slowed down with time. This observation is found for PI(BTDA-DAH) specimen films which are stretched to 50% and 200% beyond their initial length. The change in the deformation during relaxation increased with the level of strain deformation during the uniaxial stretching. The higher the level of strain deformation during the uniaxial stretching, the larger and faster the strain deformation during relaxation. However, a critical level of deformation during the stretching is found; above this level of deformation, the change in the strain deformation during relaxation decreases. The critical level of deformation is ԑ=150%.

120

FIGURE 5.5 NORMALIZED TRUE STRAIN AS A FUNCTION OF RELAXATION TIME FOR PI(BTDA-DAH) RELAXED AT 100°C FOR 10 MINUTES.

5.3.3 Optical Behavior

Figure 5.6 shows the effect of 10 minutes of relaxation on the change in birefringence behavior. The birefringence values of the specimen films are normalized to their initial strain value at relaxation time zero. This makes possible a comparison between relaxing PI(BTDA-DAH) specimen films that are stretched to different strain levels. Three trends of birefringence behavior are found. At low strain levels (ԑ=10-100%), the birefringence rapidly decreases at the beginning of relaxation; then, it slows down and maintains an almost constant level until the end

121

of relaxation. Secondly, at intermediate strain levels (ԑ=125-150%), the birefringence rapidly increases. The higher the level of strain deformation during the uniaxial stretching, the higher the change in birefringence during relaxation. Above the critical level of deformation ԑ=150%, the change in birefringence is reduced with increasing levels of strain deformation during uniaxial stretching until almost no change is observed at ԑ=250%. The similarity in the change in birefringence and strain during relaxation at intermediate and high levels of strain deformation during uniaxial stretching revealed a direct link between birefringence and strain; this is due to spontaneous deformation.

FIGURE 5.6 NORMALIZED BIREFRINGENCE AS A FUNCTION OF RELAXATION TIME FOR PI(BTDA-DAH) RELAXED AT 100°C FOR 10 MINUTES. THE VALUES INDICATE THE ENGINEERING STRAIN AT RELAXATION TIME ZERO.

122

5.4.4 Rheo-Optical Behavior

Figure 5.7 shows the real-time evolution of vertically and horizontally polarized IR absorbance of PI(BTDA-DAH) during stretching followed by 10 minutes of relaxation. The dichroic ratios (D, Do) of the PI(BTDA-DAH) is calculated from the IR absorbance at 1770 cm-1. [149] The vibrational mode at 1770 cm-1 is C=O asymmetric stretch with a transition dipole moment angle () of 0°, [150] as can be seen in Figure 5.8. The change in orientation functions (fIR) during stress relaxation is shown in Figure 5.9. The orientation function values of the specimen films are normalized to their initial strain value at relaxation time zero. This makes possible a comparison between relaxing PI(BTDA-DAH) specimen films that are stretched to different strain levels. The changes in birefringence behavior described in 5.4.3 are nearly replicated by this orientation function measurement. Three trends of change in orientation function behavior are detected. At low strain levels (ԑ=10-

100%) the change in orientation functions decrease at the beginning of relaxation; it then slows down and maintains an almost constant level until the end of relaxation.

Unlike the change in birefringence shown in Figure 5.6, the orientation function of relaxing specimens that are stretched 10% beyond their original length is similar to those that are stretch 50% beyond their original length. The similarity in the orientation functions for these specimens is due to a resolution error of the URS-FTIR instrument. At intermediate strain levels (125-150%) the orientation function rapidly increases. The level of change in the orientation function increases with the level of strain during uniaxial deformation. A critical level of strain deformation is detected;

123

above this value, the change in orientation function starts to reduce until almost no change is observed at 250%. Due to spontaneous deformation, the similarity in the change in orientation function, birefringence, and strain during relaxation reveals a direct link between the order of the polymer chains and strain.

124

Vertical IR Polarization Horizontal IR Polarization

C=O -1 C=O 1770 cm -1 1770 cm

FIGURE 5.7 REAL-TIME EVOLUTION OF IR ABSORBANCE FOR THE C=O ASYMMETRIC PEAK AT 1770CM-1 DURING STRETCHING AND RELAXATION

125

Draw axis Chain axis Transition dipole moment angle

FIGURE 5.8 PI(BTDA-DAH) CHAIN ORIENT FUNCTION AND ITS DIPOLE TRANSITION ANGLE IN A UNIAXIALLY ORIENTED POLYMER.

FIGURE 5.9 NORMALIZED ORIENTATION FUNCTION AS A FUNCTION OF RELAXATION TIME FOR PI(BTDA-DAH) RELAXED AT 100°C FOR 10 MINUTES. THE VALUES INDICATE THE ENGINEERING STRAIN AT RELAXATION TIME ZERO.

126

5.3.5 Mechano-Optical Behavior

Figure 5.10 shows the true stress-birefringence behavior of solution cast

PI(BTDA-DAH) specimen films during uniaxial stretching followed by 10 minutes of relaxation. In the previous chapter, it was found that true stress-birefringence behavior for this polymer followed the I-II-IIIc regimes during uniaxial deformation at 100°C at a stretching rate of 10mm/min. In short, in regime I during the early stage of deformation, birefringence linearly increases with stress and the stress-optical rule

(SOR) is found to be valid. In this study the stress optical constant is 8.35GPa-1. In regime II a deviation from the SOR is observed where a small increase in deformation stress results in a sharp increase in birefringence. A rapid increase in the crystallinity is observed by the end of regime II. This occurrence is due to stress-induced crystallization of the polymer chains during stretching in regime II. In regime IIIc an inverse behavior to that observed in regime II takes place: a large increase in stress results in only a small increase in birefringence. There is no significant change in the crystallinity when further stretching is undertaken in regime IIIc, where the polymer chains reach their finite extensibilities.

Martins et al. [148] studied the mechano-optical behavior of PEN specimen films during stretching followed by stress relaxation. In their study, the PEN specimen films, which follow I-II-IIIc regimes, are stretched to different stretching levels in I-II-IIIc regimes. It is found that the mechano-optical behavior of PEN specimen films during relaxation depends on the structure of the PEN films at the beginning of relaxation. Martins et al. divided the mechano-optical behavior during

127

the relaxation process into I-II-III regimes. In brief, in regime I, which is the early stage of deformation, birefringence linearly decreases with stress. At an early stage of deformation in regime II, the stress relaxation is accompanied by a decrease in birefringence, until a critical point is reached at which the birefringence starts to increase. At the intermediate stage of deformation in regime II, the stress relaxation is accompanied by no change in birefringence until a critical point at which the birefringence starts to increase. At the high stage of deformation in regime II, the stress relaxation is accompanied by increasing birefringence. In regime III, no change in birefringence is observed during stress relaxation.

Similar, but not identical, mechano-optical behavior is revealed in the three regimes for PI(BTDA-DAH). When the PI(BTDA-DAH) specimen film is stretched to strain levels within regime I (ԑ=10%), the birefringence decreases linearly with the stress on the same SOR curve. At the early stage of deformation in regime II (ԑ=50%), birefringence decreases with the stress during relaxation. In this case, the change in birefringence is slower than the change in stress. At the early intermediate stage of deformation in regime II (ԑ=100%), the birefringence continues to decrease with the stress during relaxation until a critical point at which birefringence starts to increase.

At the intermediate stage of deformation in regime II (ԑ=125%), the birefringence does not change during relaxation until a critical point which birefringence starts to increase. At the high stage of deformation in regime II (ԑ=150%), the birefringence increases during relaxation. Unlike the results found in the Martins et al. study on

PEN thin film, the birefringence of PI(BTDA-DAH) continues to increase, if slowly,

128

during regime III. At the high stage of deformation in regime III, the birefringence increases slower with relaxation.

FIGURE 5.10 THE MECHANO-OPTICAL BEHAVIOR OF PI(BTDA-DAH) DURING THE STRETCHING AND RELAXATION PROCESS. SAMPLES WERE STRETCHED BETWEEN 10% AND 250% BEYOND THEIR ORIGINAL LENGTH AT 100°C AND 10MM/MIN AND RELAXED FOR 10 MINUTES.

5.3.6 Mechano- Rheo-Optical Behavior

Figure 5.11 show true stress-orientation function behavior of solution cast

PI(BTDA-DAH) specimen films during uniaxial stretching followed by 10 minutes of relaxation. In our knowledge, this is the first time that real-time orientation

129

function has been presented as a function of stress for slow-crystallization polymers.

As with birefringence behavior, three regimes (I-II-IIIc) are revealed during uniaxial deformation as well as relaxation. In regime I (during the early stage of deformation), the orientation function linearly increases with stress. In regime II a deviation from linearity is observed in which a small increase in deformation stress results in a sharp increase in the orientation function. In regime III a large increase in stress results in a small increase in the orientation function.

When the PI(BTDA-DAH) specimen film is stretched to regime I strain levels and is followed by relaxation, the orientation function measured during relaxation decreases linearly with the stress in a way that parallels the orientation development found during stretching. At the early stage of deformation in regime II, orientation function decreases with the stress during relaxation; however, while this decrease is linear, the orientation function decreases slower than stress. At the early intermediate stage of deformation in regime II (ԑ=100%), the orientation function continues to decrease with the stress during relaxation until the critical point at which birefringence starts to increase. At the intermediate stage of deformation in regime II

(ԑ=125%), the orientation function does not change during relaxation until the critical point at which orientation function starts to increase. At the high stage of deformation in regime II (ԑ=150%), the orientation function increases during relaxation. At the early stage of deformation in regime III (ԑ=200%), the orientation function more slowly increases during relaxation. At the high stage of deformation in regime III

(ԑ=250%), the orientation function increases more slowly with relaxation.

130

0.5 I II III

0.4 2 2.5 0.3

f 1.5 0.2

0.1 1 1.25 0.5 0.1 0.0 0 2 4 6 8 10 12 14 (MPa) T

FIGURE 5.11 THE MECHANO-RHEO OPTICAL BEHAVIOR OF PI(BTDA-DAH) DURING THE STRETCHING AND RELAXATION PROCESS. SAMPLES WERE STRETCHED BETWEEN 10% AND 250% BEYOND THEIR ORIGINAL LENGTH AT 100°C AND 10MM/MIN AND RELAXED FOR 10 MINUTES.

5.3.7 Structural Studies: Effect of Stretching Ratio

Off-line experiments (x-ray and DSC) are performed on series of PI(BTDA-

DAH) specimen films that are stretched to different deformation levels followed by 10 minutes’ relaxation at 100°C. These experiments clarify the structural mechanisms responsible for the different mechano-optical behavior observed during uniaxial stretching and relaxation. Figure 5.12 shows the mechano-optical behavior of a series of PI(BTDA-

DAH) specimen films, together with WAXD patterns and their level of crystallinity.

131

Throughout stretching and relaxation of PI(BTDA-DAH) specimen films in regime I, the polymer remains amorphous, as indicated by a broad halo in the WAXD pattern. The same observation is found for the PI(BTDA-DAH) specimen film which is stretched to an early stage of deformation in regime II, at a strain level of 50%. When the level of stretching increases to 100% beyond the PI(BTDA-DAH) specimen film’s original length, the polymer film remains amorphous, as indicated by its low crystallinity levels and by a broad halo in the WAXD pattern. However, after 10 minutes of relaxation, the level of crystallinity significantly increases. Furthermore, a weak equatorial peak starts to appear, coupled with a streak diffraction along the meridian direction. These observations indicate that during 10 minutes of relaxation, the polymer chain starts to develop a disordered crystalline structure which appears periodically throughout the polymer chain. Increasing the uniaxial stretching to the level of deformation of 125% increases the crystallinity.

However, no evidence of ordered structure is observed in the WAXD pattern. During 10 minutes of relaxation, the level of crystallinity continues to increase. Furthermore, a sharp equatorial peak appears, coupled with 4 diffraction streaks along the meridian direction.

These observations indicate that, during 10 minutes of relaxation, strain-induced crystallization occurs. Moreover, this indicates that a disorder in the crystalline structure occurs regularly throughout the polymer chain. A PI(BTDA-DAH) specimen which is uniaxially stretched to 150% beyond its original length shows high levels of crystallinity.

The WAXD pattern shows sharp equatorial peaks, which indicate high levels of chain orientation. Moreover, 3 substantial streak diffractions are observed along the meridian direction. After 10 minutes of relaxation, the polymer film has higher levels of crystallinity.

The WAXD pattern shows very sharp equatorial peaks, which indicates a very high level

132

of chain orientation. The number of streak diffractions along the meridian direction increases to 7, which indicates an increased regularity of the periodic disordered crystalline structure. Throughout regime III, at strains of 200% and 250% beyond their original lengths, a high level of crystallinity is observed. The WAXD pattern shows very sharp equatorial peaks which indicates a high level of chain orientation. Moreover, along the meridian direction 7 substantial streak diffractions are observed. During 10 minutes of relaxation, the crystallinity slightly increases. No change in the WAXD patterns are observed after relaxation. Throughout all of the deformation levels, the absence off- equatorial diffraction spots in the WAXD patterns suggest that a three-dimensional crystalline order is absent.

133

휙 = 15.7% 휙 = 8%Stretched 휙 = 8.0% 휙 = 14.1% 휙 = 15.6%

+ 10 min 휙 = 11.4%

휙 = 3.4%

휙 = 4.84% 휙 = 9.4%

휙 = 0.6%

휙 = 8.0%

휙 = 0.5%

휙 = 0.3%

휙 = 4.6%

휙 = 0.3% 휙 = 3.3% 휙 = 0.3% 휙 = 0.5%

Stretched

휙 = 0.5% FIGURE 5.12 MECHANO-OPTICAL BEHAVIOR OF PI(BTDA-DAH) DURING THE STRETCHING AND RELAXATION PROCESS WITH CORRESPONDING WAXD PATTERN. SAMPLES ARE STRETCHED AT 100°C AND 10MM/MIN AND RELAXED FOR 10 MIN. THE VALUES INDICATED IN THE PICTURES ARE REFERRING TO THE LEVEL OF DEFORMATION DURING UNIAXIAL STRETCHING AND TO THE LEVELS OF CRYSTALLINITY.

134

DSC thermograms of relaxed and unrelaxed stretched PI(BTDA-DAH) are presented in Figure 5.13. The cold crystallization temperature and the area under melting peaks are affected by the level of deformation and by the relaxation. A comparison of only the stretched specimens indicates that the cold crystallization temperatures decreased.

These changes in the cold crystallization temperature are a result of decreases in entropy; this entropy is due to a higher arrangement of the polymer chains in the amorphous region, as well as an increase in their crystallinity. Unlike PET, PEN, and other cold crystallization polymers, PI(BTDA-DAH) continues to crystallize by strain-induced crystallization in regime III, as can be seen in Figure 5.14. As a result, the level of birefringence and the orientation function increases in regime III (as shown in Figure 5.10 and 5.11).

The cold crystallization temperature peak is lower in almost all of the relaxed specimens than it is in the stretched films of the same level of deformation. This means that the entropy is decreasing during relaxation due to an increase in the order of the amorphous region. In a relaxed film which is stretched to 150% beyond its original length, the cold crystallization peak disappears. This disappearance of the cold crystallization peak indicates a decrease in entropy due to an even higher arrangement of the polymer chains in the amorphous region than is witnessed in the lower stretching levels.

During relaxation, the polymer chain starts to relax, translate and change conformation, in order to reduce the chain’s energy state (entropy). As a result of this behavior, some of the polymer chains become parallel to each other, which is a prerequisite of strain-induced crystallization. This crystallization process increases the orientation of the polymer chains in the amorphous region.

135

휀 = 0.1 휀 = 0.1 (푅) 휀 = 0.5 휀 = 0.5 (푅) 휀 = 1 휀 = 1 (푅) 휀 = 1.25 휀 = 1.25 (푅) 휀 = 1.5 휀 = 1.5 (푅) 휀 = 2 휀 = 2 (푅) 휀 = 2 .5 휀 = 2 .5(푅)

FIGURE 5.13 DSC THERMOGRAMS OF STRETCHED AND RELAXED PI(BTDA-HDA) SAMPLES WHICH HAVE BEEN STRETCHED AT DIFFERENT DEFORMATION LEVELS AT 100°C AND A RATE OF 10MM/MIN AND THEN RELAXED FOR 10 MIN.

There are two deformation phenomena which control the development of crystallinity in the PI(BTDA-DAH) specimen. Firstly, the crystallinity increases during stretching because of strain-induced crystallization. Secondly, the crystallinity increases during the relaxation stage due to spontaneous deformations as can be seen in Figure 5.14.

In Figure 5.15, the crystallinity of stretched specimens is plotted as a function of true

(Hencky) strain and compared with that of specimens which have been stretched and then relaxed for 10 minutes. The crystallinity only increases during relaxation if the specimen is stretched 125% beyond its original length or more. This increase accompanies an

136

increase in the Hencky strain. Several associated behaviors are found. During the early strain deformation in regime II, there is almost no change in the Hencky strain and crystallinity after 10 minutes of relaxation. At the early strain deformation in the intermediate stage of regime II, the crystallinity increases with the strain. A rapid increase in crystallinity is observed in specimens which are stretched at the intermediate stage of regime II. Furthermore, after 10 minutes of relaxation, the development of the crystallinity approaches the maximum for this polymer. The crystallinity is similar to that which develops after 10 minutes of relaxation in specimens that are relaxed in regime III. The change in the crystallinity of specimens which were stretched to the intermediate level of regime II and relaxed for 10 minutes and 1 hour was less than 2%, as can be seen in Figure

5.16.

FIGURE 5.14 PERCENTAGE OF CRYSTALLINITY OF PI(BTDA-DAH) AS A FUNCTION OF TRUE STRESS. THE CIRCLES REPRESENT STRETCHED SPECIMENS AND THE TRIANGLES REPRESENT SPECIMENS THAT HAVE BEEN STRETCHED AND RELAXED. ALL THE SAMPLES ARE STRETCHED AT DIFFERENT STRAIN DEFORMATIONS DURING UNIAXIAL DEFORMATION AT 100°C AND AT A RATE OF 10MM/MIN.

137

FIGURE 5.15 PERCENTAGE OF CRYSTALLINITY OF PI(BTDA-DAH) AS A FUNCTION OF TRUE (HENCKY) STRAIN. THE CIRCLES REPRESENT STRETCHED SPECIMENS AND THE TRIANGLES REPRESENT SPECIMENS THAT HAVE BEEN STRETCHED AND RELAXED. ALL THE SAMPLES ARE STRETCHED AT DIFFERENT STRAIN DEFORMATIONS DURING UNIAXIAL DEFORMATION AT 100°C AND AT A RATE OF 10MM/MIN.

5.3.8 Structural Studies: Effect of Relaxation Time

Off-line experiments (x-ray and DSC) coupled with real-time rheo-mechano- optical behavior are performed on a series of PI(BTDA-DAH) specimen films which are stretched 100% and 150% beyond their original length and relaxed for varying relaxation times (0 min,1 min,10 min,60 min) at 100°C. These experiments clarify the structural mechanisms responsible for the different rheo-mechano-optical behaviors observed during uniaxial stretching and relaxation. Figure 5.16 shows the mechano-optical behavior, mechano-orientation function behavior of PI(BTDA-DAH) relaxed specimen films which

138

were stretched 150% beyond their original length, the WAXD patterns and DSC thermograms. At this level of deformation, the strain-induced crystallization occurs during stretching. Although cold crystallization temperatures are still observed, the x-ray pattern shows a high order level of the polymer chains. During 1 minute of relaxation, the orientation of the polymer chains in the matrix increases. The birefringence and orientation function increases during the relaxation. Furthermore, the cold crystallization peaks decrease. The polymer chains’ orientations continue to increase with the relaxation time.

After 10 minutes of relaxation, the birefringence as well as the orientation function increase dramatically in comparison to the increase witnessed after 1 minute of relaxation. A high order of orientation can be observed in the x-ray pattern of the specimens that were relaxed for 10 minutes. An increase in crystallinity is also indicated by an increase in the area of the melting peak and a decrease in the area of the cold crystallization peak. The effect of more relaxation time (10 minutes’ relaxation vs. 60 minutes’ relaxation) was minor; there was very little increase in polymer orientation and crystallinity. However, the melting peak of the specimen that relaxed for an hour is sharper, which indicates an improvement in the crystals during 1 hour of relaxation.

Figure 5.17 shows the mechano-optical behavior, mechano-orientation function behavior of PI(BTDA-DAH) relaxed specimen films which were stretched 100% beyond their original length, WAXD patterns and DSC thermograms. The sample described by

Figure 5.15 has been stretched to the early stage of deformation in regime II. Therefore, strain-induced crystallization does not occur during stretching and the polymer specimen is isotropic, as can be seen in the x-ray pattern and the DSC thermograph. During the first minute of relaxation, the film anisotropy decreases, as can be seen in the birefringence and

139

orientation function behaviors. The polymer chains attempt to rearrange themselves in coil formation in order to reduce the energy state in the polymer matrix. In this case, the polymer chains are not parallel; therefore, strain-induced crystallization does not occur.

The x-ray pattern suggests that the polymer specimen is still isotropic. During 10 minutes of relaxation the birefringence continues to decrease while the orientation function sharply increases. These results may suggest that the orientation function which is calculated for asymmetric C=O stretching (1770 cm-1) is more associated with the amorphous region than is typical of the crystalline regions. The WAXD pattern indicates an initial stage of order of the polymer chains’ orientations. At the end of 10 minutes of relaxation, the birefringence starts to increase. The effect of more relaxation time (10 minutes’ relaxation vs. 60 minutes’ relaxation) is significant. The amount of crystallinity increases, and as a result the birefringence rapidly increases. Furthermore, a high order of orientation can be observed in the WAXD pattern.

140

FIGURE 5.16 DETAILED MECHANO-OPTICAL BEHAVIOR, ORIENTATION FUNCTION BEHAVIOR WAXD PATTERNS, AND DSC THERMOGRAMS OF PI(BTDA-HDA) STRETCHED TO 150% BEYOND THEIR ORIGINAL LENGTH AND RELAXED FOR 1 H.

141

FIGURE 5.17 DETAILED MECHANO-OPTICAL BEHAVIOR, ORIENTATION FUNCTION BEHAVIOR WAXD PATTERNS, AND DSC THERMOGRAMS OF PI(BTDA-DAH) STRETCHED TO 100% BEYOND THEIR ORIGINAL LENGTH AND RELAXED FOR 1 H.

142

5.3.9 Structural Evolution

A schematic representation of the structural evolution of PI(BTDA-DAH) after its uniaxial stretching to its early level of deformation in regime II based on the above results and discussions is provided in Figure 5.18. At the beginning of deformation, the polymer is in an amorphous state. The polymer chains are unoriented and this order remains until the critical point between regime I and regime II. During early deformation in regime II, the orientation of the polymer chain begins. The polymer chain starts to organize and orient between its chain entanglement nods. During this deformation stage the chains are oriented; however, they are not aligned parallel to each other. As a result, strain-induced crystallization is not possible, as can be seen by the amount of crystallinity and the WAXD pattern in Figure 5.15. During relaxation at this stage, the polymer chain tries to rearrange itself in a coil formation, which reduces the energy state in the polymer matrix. However, the recovery is limited due to the chain entanglement nods. Soon after relaxation begins for a sample that has been stretched to the early stage of regime II, the polymer chains start to recover and rearrange themselves, which decreases the anisotropy of the polymer matrix.

Therefore, a decrease in the birefringence and orientation function are observed. This recovery increases with the relaxation time as a result of a decrease in the birefringence. In order to minimize the entropy, the polymer chains which relaxed rearrange so that they are parallel to each other. This causes the material to crystallize.

143

FIGURE 5.18 SCHEMATIC REPRESENTATION THE STRUCTURAL EVOLUTION OF PI(BTDA-DAH) DURING RELAXATION AFTER UNIAXIAL STRETCHING.

144

5.3.10 Dielectric properties

Figure 5.19 shows the dielectric constant of PI(BTDA-DAH) relaxed specimen films which are stretched 150% beyond their original length. In this study, the effect of frequency and relaxation time on dielectric constant is investigated. The dielectric constant of PI(BTDA-DAH) is sensitive to frequency, and relaxation time. First, the dielectric constant slightly decreases at higher frequencies. Second, the dielectric constant increases at higher relaxation time. As a result of strain-induced crystallization, the total polarizability of the polymer chain increases in the thickness direction therefore, the dielectric constant increases (more details is provided in chapter VI).

FIGURE 5.19 THE EFFECT OF RELAXATION TIME ON THE DIELECTRIC CONSTANT OF PI(BTDA-DAH) RELAXED SPECIMEN FILMS WHICH WERE STRETCHED 150% AT DIFFERENT FREQUENCIES. THE TEST WAS MADE AT 25°C.

145

The D-E loops of PI(BTDA-DAH) relaxed specimen films which were stretched

150% beyond their original length shown in Figure 5.20. The maximum voltage which is applied without making a catastrophic failure (electrical breakdown), increases with the relaxation time. This indicates that higher level of crystallinity increases the dielectric breakdown of PI(BTDA-DAH) films.

FIGURE 5.20 THE EFFECT OF RELAXATION TIME ON THE D-E LOOPS OF PI(BTDA-DAH) RELAXED SPECIMEN FILMS WHICH WERE STRETCHED 150% AT DIFFERENT FREQUENCIES. THE TEST WAS MADE AT 25°C.

5.4 Conclusions

The relaxation behavior of uniaxially stretched PI(BTDA-DAH) thin films in their rubbery state is studied. The rheo-mechano-optical behavior during relaxation is found to depend on the amount of deformation undergone during uniaxial stretching. Three regimes of mechano-optical behavior and mechano-rheo behavior are revealed during the uniaxial 146

stretching and relaxation. In regime I, during the early stage of deformation, birefringence and orientation function linearly increase with stress during the stretching, and birefringence and orientation function linearly decrease with stress during relaxation. In regime II a deviation from the linearity is observed during stretching in which a small increase in deformation stress results in a sharp increase in birefringence. During relaxation, the birefringence and orientation function behaviors depend on the level of the stress deformation undertaken during uniaxial stretching. For instance, at the early stage of deformation in regime II, birefringence and orientation function decrease with the stress during relaxation. At the early intermediate stage of deformation, the birefringence and orientation function continue to decrease with the stress during relaxation until a critical point at which birefringence and orientation function start to increase. At the intermediate stage of deformation in regime II, the stress relaxation is accompanied by no change in birefringence until a critical point at which the birefringence starts to increase. At the high stage of deformation in regime II, the stress relaxation is accompanied by increasing birefringence and orientation function. In regime IIIc, a negative deviation of birefringence and orientation function from linearly are observed during uniaxial stretching. During relaxation in regime III, an increase in birefringence and orientation function are observed.

The increase in birefringence and orientation function is a result of a spontaneous deformation process which occurs during relaxation. The level of crystallinity and the perfection of the crystals increases as a results of an increase in the relaxation time allotted for specimens which are stretched in regimes II and III. The relaxation time has a critical role on the breakdown strength. Higher relaxation time improves the maximum breakdown strength of the PI(BTDA-DAH) films.

147

CHAPTER VI

REAL-TIME MECHANO-OPTICAL BEHAVIOR AND STRUCTURAL EVOLUTION

OF PI (BTDA-DAH) DURING BIAXIAL DEFORMATION.

This chapter focuses on the real-time mechano-optical behavior of PI(BTDA-DAH) during two types of biaxial deformation: simultaneous and sequential. The mechano-optical behavior of the polymer is studied at 90°C at different strain levels, deformation rates and types of biaxial deformation. At this temperature, the polymer is in its rubbery state. This study is performed with an instrumented system that combines real-time spectral birefringence measurements with true stress and true strain during biaxial deformation.

Off-line characterization techniques (DSC, WAXD, and Abbe refractometer), are used to identify the films’ structural origins. IMASS Time Domain Dielectric Spectrometers with novocontrol GmbH concept 40 broadband dielectric spectrometers are used to evaluate the dielectric properties of the biaxially stretched films. The dielectric constant of the

PI(BTDA-DAH) specimens depends on the level of their biaxial deformation. A PI(BTDA-

DAH) specimens which are stretched to higher stretching ratios have lower dielectric constants. The Abbe refractometer and the real-time spectral birefringence measurements revealed that the phenyl groups in the PI(BTDA-DAH) specimens’ backbone chains began to parallel orient to the PI(BTDA-DAH) film surface plane at higher biaxial stretching

148

ratios. This chain arrangement decreases the total polarizability of the PI(BTDA-DAH) specimen's thickness. As a result, the dielectric constant of the film's thickness decreases.

6.1 Introduction

Metallized biaxially oriented polypropylene (BOPP) is currently the most popular material used in polymeric dielectric capacitors, although it is not ideal for this application.

[78,79] It suffers from a low dielectric constant, a low energy density at breakdown, and a low operating temperature. [79-81,113] However, its self-healing characteristics, low loss factor, and high dielectric strength compensate for its weaknesses. [79,80]

The conditions under which thin PP films are manufactured impacts the final state of their polymer chain orientation and crystallinity; in turn, these factors largely dictate the dielectric breakdown of PP films. For instance, the dielectric breakdown of unoriented PP decreases when the spherulite diameter is increased. The larger the size of the spherulitic structure, the lower the film’s maximum breakdown strength. (120) When stretched, PP’s spherulitic structure breaks and it converts to a fibrillar one. This increases its dielectric breakdown. [88,121]

Industries use the tenter-frame machine for capacitors because the breakdown strength of the BOPP made with this machine is higher than that made with film blowing; the BOPP that results from film blowing has a distinctly different shish kebab morphology that includes significant levels of chain-folded lamellae. [53] Furthermore, the final film

149

thickness that can be achieved using the tenter-frame machine is lower than that possible with film blowing: 3 μm as compared to 10 μm. [53]

PI(BTDA-DAH) was synthesized by Rui et al. This polymer exhibits dielectric constants of 3.57 and a breakdown strength of 812 (V/ μm) when it is cast by a solution processing method. [119] The structural evolution, rheo-mechanical-optical behavior, and dielectric properties of this polymer during uniaxial stretching followed by relaxation are discussed in the previous chapter. This chapter focuses on the real-time structural development of a PI(BTDA-DAH) during simultaneous and sequential biaxial stretching and its effect on the specimen’s dielectric properties.

6.2 Experimental Details

6.2.1 Materials

PI(BTDA-DAH) polymer is synthesized by the two-step polycondensation reaction shown in Figure 6.1. First, poly(amic acid) is synthesized by a reaction between M-Cresol

(which acts as the solvent), 3,3′,4,4′-benzophenone tetracarboxylic dianhydride (BTDA),

1,6-diaminohexane (DAH) monomers and isopropylamine (which acts as the end-capping reagent). The reaction is carried out at 30°C for 24 hours. In the second step, PI(BTDA-

DAH) is formed by the imidization of polyamic acid. This imidization reaction is carried out at 165-175°C for 6 hours.

150

30°C M-Cresol 24 hours

R.T

165°C-175°C M-Cresol

FIGURE 6.1 SYNTHESIS OF PI (BTDA-DAH) POLYMER.

6.2.2 Sample Preparation

PI(BTDA-DAH) 13 wt.% solution in M-Cresol is made. The solution is filtrated through a 0.45 µm PTFE filter. Unoriented films are prepared by a casting procedure. The solution is cast on a PET substrate (Mylar) in a roll to roll casting line by using a 6'' wide casting doctor blade. The films are dried for 48 hours at room temperature and then at 60°C for another 12 hours. In order to peel the PI(BTDA-DAH) films from the PET substrate, the films are kept in DI water for 3½ hours. A final thickness of 28 µm is obtained. Square specimens are cut for further characterization.

151

6.2.3 Characterization

PI(BTDA-DAH) specimens in their rubbery states are simultaneously and sequentially stretched at 90°C with varying stretching rates of 5 mm/min, 50 mm/min, and

500 mm/min; the stretching ratios investigated are 1.2X1.2, 1.4X1.4, 1.6X1.6, and 1.8x1.8.

A real-time mechano-optical measurement platform equipped with a visible light spectrometer is used to study the deformation behavior of PI(BTDA-DAH) during the biaxial stretching. Details of the measurement system are discussed elsewhere. [151] In brief, a specimen is clamped to a biaxial stretching machine enclosed by a heating chamber.

The film is held by 4 sets of pneumatic clamps. The motions of the pneumatic clamps are controlled by 2 motors; therefore, the pneumatic clamps can move independently in x and y directions. As a result, the polymer film can be stretched simultaneously and sequentially.

Two 200 N load cells are connected to two pneumatic clamps located in the middle of the pneumatic clamp sets; these load cells measure the forces that are applied through the stretching in the MD and TD. A CCD camera measures the true strain during the stretching experiment by detecting the movement of 24 dots which were marked prior to the experiment. By assuming incompressibility (퐿푥0퐿푦0푑0 = 퐿푥푡퐿푦푡푑푡) and by measuring the change of the specimen length, the real-time change in the polymer thickness is calculated by the following equation:

퐿푥0×퐿푦0×푑0 푑푡 = (6.1) 퐿푥푡×퐿푦푡

152

where 퐿 is the length and d is the film’s thickness. The subscripts o and i indicates initial value and real-time value respectively while x and y indicates the direction of the deformations. By knowing the change in the specimen’s thickness, the change in the specimen’s cross-sectional area can be obtained. As a result, the true stress is calculated by the following equation:

퐹푡 퐹푡× 퐿푥푡×퐿푦푡 휎푡 = = (6.2) 퐿푖푡×푑푡 퐿푖푡×퐿푥0×퐿푦0×푑0 where F and 휎 are force and the true stress respectively. The subscript i represents the direction of stretching (x or y). Two visible wavelength light sources are used to measure the optical retardation () with resolution of ± 2nm through the polymer film. The in-plane birefringence (n12) and out-of-plane birefringence (n23) are calculated using the following equation:

훤0 ∆푛12 = (6.3) 푑푡

sin2 ∅ 훤 −훤 √(1− ) 1 0 ∅ 푛̅2 ∆푛23 = − ( sin2 ∅ ) (6.4) 푑0 푛̅2

Where  and 푛̅ are retardation and average refractive index respectively. ∅ is the tilted angle which the 훤∅ is measured; for this experiment the tilted angle is 45°.

WAXD measurements are performed using a Bruker AXS D8 Goniometer. The instruments uses CuKα radiation (λ = 0.1542 nm). WAXD diffractograms are used to measure any changes in the crystalline structure of the PI(BTDA-DAH) upon its stretching.

153

A TA Instruments differential scanning calorimeter (DSC Q-200) is used to measure the glass transition, cold crystallization range, melting point (Tm) and heat of fusion (Hf) of PI(BTDA-DAH) films. Before the DSC measurements, temperature calibration is performed using an Indium standard (Tm = 156.6C). A heating rate of

∆퐻푓−∆퐻푐 휙푚 = (6.5) ∆퐻푃퐼

where Hf is the area under the DSC melting endotherm, Hc is the area under the DSC cold crystallization exotherm, and HPI is the enthalpy of fusion of a pure PI crystal which is 144.14 J/g. [123]

A Bellingham + Stanley 60 Abbe refractometer is used to measure the principal refractive index (MD, TD, and ND) of the PI(BTDA-DAH) specimen films before and after they have been stretched. These measurements are carried out using a white light source with a band-pass filter (589.3 nm) to generate the monochromatic light.

100°C, 125°C, and 150°C are applied to measure the dielectric constant and loss.

Breakdown strength measurements are performed using a linear voltage ramp of

300 V/sec generated by a resistor capacitor (RC) circuit on 1.4cm X 1.4cm specimens. The breakdown voltage of the sample is read from a peak-holding voltmeter.

154

6.3 Results and Discussion

6.3.1 Mechanical Behaviors

Figure 6.2 and 6.4 shows the effect of true (Hencky) strain over time on solution cast PI(BTDA-DAH) films during equal simultaneous and sequential biaxial deformation at different stretching rates and ratios. In these studies, the polymer films are heated to

90°C and isothermally maintained at that temperature for 10 minutes. In the data illustrated in Figure 6.2, the polymer films are sequentially stretched up to 1.8 X 1.8 times their original length at stretching rates of 50mm/min and 500 mm/min. Basically, sequential biaxial stretching is a combination of two UCW steps. First UCW stretching of the polymer specimen occurs in the MD. Secondly, UCW stretching of the polymer specimen occurs in the TD. During the first step, the MD Hencky strain increases. Unlike what is expected, the TD Hencky strain decreases at this step (negative values of strain are observed), although there is no movement of the pneumatic clamps in the TD. This negative strain value results from local deformation. During the UCW in the MD, the distance between the dots marked on the film surface increases in the MD while they slightly retract in the

TD (as can be seen in Figure 6. 3). As a result, a negative value of Hencky strain in the TD is observed. During the second step of the sequential biaxial stretching, the TD Hencky strain increases while the MD Hencky strain decreases. Again, during the UCW in the TD the dots' distance from each other increases in the TD while they slightly retract in the MD.

Furthermore, the final Hencky strains in the MD and the TD are found to be unequal. At a low stretching ratio the MD Hencky strain is higher than the TD Hencky strain. When the

155

stretching ratios increase the differences between the MD Hencky strain and the TD

Hencky strain reduce until a critical point at which the value of the TD Hencky strain becomes higher than the MD Hencky strain. This inequality in the final MD and TD

Hencky strain values is found only in sequential biaxial stretching. This observation has not found for PI(BTDA-DAH) films which were equal simultaneous biaxial stretched as can be seen in Figure 6.4.

500 mm/min 1.8X1.8 End of USW in MD and 50 mm/min beginning of USW in TD TD 1.8X1.8

1.6X1.6 MD TD 1.4X1.4

1.2X1.2

FIGURE 6.2 EVOLUTION OF HENCKY STRAIN DURING SEQUENTIAL BIAXIAL STRETCHING AT 90°C AND DIFFERENT RATES AND STRETCHING RATIOS.

156

UCW

FIGURE 6. 3 CHANGE IN THE 24 DOTS DURING UCW.

FIGURE 6.4 EVOLUTION OF HENCKY STRAIN DURING SIMULTANEOUS BIAXIAL STRETCHING AT 90°C AND DIFFERENT RATES AND STRETCHING RATIOS.

157

6.3.2 Optical Behaviors

Figure 6.5 shows the evolution in the in-plane and the out-of-plane birefringence behavior of solution cast PI(BTDA-DAH) films during equal simultaneous biaxial deformation over time (logarithmic time) at different stretching rates and ratios. The in- plane birefringence stays almost the same during the simultaneous biaxial stretching. This observation indicates that PI(BTDA-DAH) specimen stays almost isotropic in the MD-TD plane throughout the equal simultaneous biaxial stretching. Unlike the in-plane birefringence behavior, the out-of-plane birefringence behavior does not stay constant or almost constant. It increases through all the stretching experiment. This observation indicates that the anisotropy in the MD-ND and TD-ND planes increases throughout the equal simultaneous biaxial stretching. In equal simultaneous biaxial deformation, the in- plane birefringence which describes the differences in the refractive indices in the MD and

TD planes (∆푛12 = 푛푀퐷 − 푛푇퐷) is close to zero. Therefore, it is possible to assume that refractive indices in the MD and TD planes are equal (푛푀퐷 = 푛푇퐷). The out-of-planes birefringence which described by the differences in the refractive indices in the MD and

ND planes and the TD and ND planes (∆푛13 = 푛푀퐷 − 푛푁퐷, ∆푛23 = 푛푇퐷 − 푛푁퐷) are not equal to zero; therefore, the ND refractive index is not equal to those of the MD and TD

(푛푀퐷 = 푛푇퐷 ≠ 푛푁퐷). Because the summation of the birefringence planes is zero (∆푛12 +

∆푛23 + ∆푛13 = 0), the out-of-plane birefringence are equal (∆푛23 = ∆푛13). The increase in out-of-plane birefringence indicates that ND refractive index decreases to lower values than the MD and TD (푛푀퐷 = 푛푇퐷 > 푛푁퐷 ) throughout the equal simultaneous biaxial stretch. During the biaxial stretching process, the aromatic rings (phenyl group) in the

158

PI(BTDA-DAH) specimen’s backbone become parallel to the specimen’s surface, which reduces the total polarizability in the ND. Therefore, the ND refractive index decreases.

Furthermore, the degree of this alignment increases with higher stretching ratios. Similar observations were found for PET during equal simultaneous biaxial stretching. [17,152]

Although the PI(BTDA-DAH) specimen is heated and stretched in its rubbery state at high stretching rates, the relaxation phenomenon is suppressed. As a result, the polymer chains do not have time to relax; therefore, a higher level of parallel alignment of phenyl groups occurs. As a result, the anisotropy increases, as indicated by the higher birefringence shows in Figure 6.5.

500 mm/min

1.8X1.8

50 mm/min 1.8X1.8 1.6X1.6 1.4X1.4 5 mm/min

1.2X1.2

1.8X1.8

∆풏ퟏퟐ

FIGURE 6.5 EVOLUTION OF THE IN-PLANE AND OUT-OF-PLANE BIREFRINGENCE DURING SIMULTANEOUS BIAXIAL STRETCHING AT 90°C, DIFFERENT RATES AND STRETCHING RATIOS.

159

Figure 6.6 shows the evolution in the in-plane and the out-of-plane birefringence behaviors of solution cast PI(BTDA-DAH) films during equal sequential biaxial deformation at different stretching rates and ratios. During the first part of sequential biaxial deformation, UCW in the MD, the in-plane birefringence increases. This indicates that the PI(BTDA-DAH) chains become more orient in the MD. As a result, the anisotropy in the MD-TD plane increases. Unlike the in-plane isotropic behavior, the out-of-plane isotropic behavior shows higher sensitivity to the UCW stretching in the MD. As a result, the out-of-plane birefringence values are a magnitude higher than the in-plane ones. As can be seen in Figure 6.6, the out-of-plane birefringence decreases during the UCW stretching in the MD. This indicates that the phenyl groups in the PI(BTDA-DAH) specimen’s backbone become perpendicular to the specimen surface, which increases the total polarizability in the ND. The degree of this alignment increases at higher stretching ratios.

Although the birefringence in the MD-ND (∆푛13) and TD-MD (∆푛23) planes are not equal

( ∆푛13 = ∆푛12 + ∆푛23 ), similar behaviors are observed. This indicates that the total polarizability of PI(BTDA-DAH) is affected mainly by the alignment of the phenyl groups in the polymer chains. During the second stretching step of the equal sequential biaxial stretching, the polymer specimen is UCW stretched to the TD, and the in-plane birefringence slightly decreases. This indicates that the PI(BTDA-DAH) chains are oriented more to the TD. The MD-TD plane birefringence at the end of this process is close to zero, which indicates that the MD-TD plane is isotropic. Unlike the in-plane, the out-of- plane is anisotropy at the end of the UCW stretching in the TD. The out-of-plane birefringence increases throughout the stretching. This indicates that the alignment of the polymer chains with the TD is accompanied by the rotation of the phenyl groups; they

160

become parallel with the polymer specimen’s surface. Therefore, the total polarizability in the ND decreases, which increases the out-of-plane birefringence. This behavior increases with higher stretching ratios. In order to get almost equal planes isotropic (∆푛12 ≈ ∆푛23 ≈

∆푛13 ≈ 0), the PI(BTDA-DAH) specimen needs to be stretched in an un-equal sequential biaxial stretching procedure.

50 mm/min 1.8X1.8 1.6X1.6 500 mm/min 1.8X1.8 1.4X1.4 1.2X1.2

End of USW in MD and beginning of USW in TD

FIGURE 6.6 EVOLUTION OF THE IN-PLANE AND OUT PLANE BIREFRINGENCE DURING SEQUENTIAL BIAXIAL STRETCHING AT 90°C AND DIFFERENT RATES AND STRETCHING RATIOS.

161

Figure 6.7 shows the refractive indices in three principal axes of an equal simultaneous biaxially stretched PI(BTDA-DAH) specimens as a function of draw ratios

( 휆푀퐷 = 휆푇퐷) . The refractive indices 푛푀퐷 and 푛푇퐷 slightly increase during the simultaneous biaxial stretching. Furthermore, their values increase at higher stretching ratios. Although 푛푇퐷 is slightly higher than 푛푀퐷, their values are similar. Therefore, the in-plane birefringence during simultaneous biaxial stretching is slightly decreased. Unlike the refractive indices in the MD and the TD, the refractive index in the ND dramatically decreases during the simultaneous biaxial stretching. Similar behavior was observed in the past for PET. [17,152] The phenyl groups in the PI(BTDA-DAH) specimen’s backbone become parallel to the specimen’s surface, which reduces the refractive index values in the

ND.

FIGURE 6.7 REFRACTIVE INDICES OF PI(BTDA-DAH) FILMS SIMULTANEOUS BIAXIAL STRETCHED TO A DIFFERENT STRETCHING RATIOS AT 90°C AND 50 MM/MIN.

162

Figure 6.8 shows the refractive indices in three principal axes of sequential biaxially stretched PI(BTDA-DAH) specimens as a function of product of draw ratios in the MD and the TD (휆푀퐷×휆푇퐷). For constant 휆푇퐷, and an increases of 휆푀퐷 cause an increasing in 푛푀퐷, 푛푁퐷and decreasing in 푛푇퐷. Unlike the stretching ratio conditions constant 휆푇퐷 and increase of 휆푀퐷, a constant 휆푀퐷 and increase of 휆푇퐷 cause an increasing in the 푛푇퐷 and decreasing in 푛푀퐷 and 푛푁퐷. Basically, the refractive index in ND decreases with increases in 휆푀퐷×휆푇퐷. Furthermore, the final values of the refractive index in the ND for equal sequentially biaxially stretched PI(BTDA-DAH) specimens are lower than those of equal simultaneously biaxially stretched PI(BTDA-DAH) specimens; the refractive indices in the MD and TD, however, are similar. This indicates that during the sequential biaxial stretching, the phenyl group becomes more parallel aligned to the film surface than during equal simultaneous biaxial stretching.

163

1.8X1 1.6X1

1.8X1.2 1.4X1 1.6X1.2 1.4X1.2 1.2X1 1.8X1.8

1.8X1.4 1.2X1.2 1.4X1.4 1.6X1.4 1.8X1.6

1.6X1.6

1.8X1.8

FIGURE 6.8 REFRACTIVE INDICES OF PI(BTDA-DAH) FILMS SEQUENTIAL BIAXIAL STRETCHED TO A DIFFERENT STRETCHING RATIOS AT 90°C AND 50 MM/MIN.

6.3.3 Strain Optical Behavior.

Off-line (x-ray and DSC) and real-time birefringence experiments are performed on a series of PI(BTDA-DAH) specimen films that are equally simultaneously and sequentially biaxially stretched at 90°C a with varying stretching ratios up to 1.8 X 1.8 and rates up to 500mm/min. Figure 6.9 shows the strain-optical behavior of series of equal simultaneously stretched PI(BTDA-DAH) specimen films, together with their WAXD patterns and their levels of crystallinity as determined by the DCS. The in-plane strain- optical behavior stays almost the same during the simultaneous biaxial stretching. An

164

almost linear dependency between strain and in-plane birefringence is observed throughout the experiment. Unlike the in-plane strain-optical behavior, the out-of-plane strain-optical behavior increases during the stretching. Furthermore, the strain-optical behavior depends on the stretching rates employed. At low stretching rates (i.e. 5 mm/min) the birefringence behavior slowly increases with strain until a critical point in stretching ratios beyond which the birefringence rapidly increases. At higher stretching rates (i.e. 500 mm/min) the birefringence rapidly increases until it starts to level off. The relaxation is suppressed at high stretching rates; therefore, the polymer relaxation which is necessary for strain- induced crystallization cannot take place. Furthermore, the DSC thermograph confirms that strain-induced crystallization takes place only at lower stretching rates. The total crystallinity at the end of simultaneous biaxial stretching at 500mm/min is less than 1%.

Relaxation occurs more at low stretching rates than it does at high stretching rates.

Therefore, a change in the shape of the strain-optical curve is observed. Furthermore, the amount of crystallinity increases (i.e at a stretching ratio of 1.8X1.8 and a stretching rate of 5mm/min the crystallinity is 7%). The WAXD patterns reveal that the in-plane (MD-

TD) stays isotropic when stretched at different stretching rates and ratios while the out of planes (MD-ND, TD-ND) obtains some order in the polymer chain.

165

=1%

=3.2%

=7%

FIGURE 6.9 STRAIN OPTICAL BEHAVIOR OF SIMULTANEOUS BIAXIAL STRETCHED PI(BTDA-DAH) FILMS TO A DIFFERENT STRETCHING RATIOS AND RATES AT 90°C.

Figure 6.10 shows the strain-optical behavior of a series of sequential PI(BTDA-

DAH) specimen films together with their WAXD patterns and levels of crystallinity as determined by the DCS. The birefringence curves are plotted as a function of Hencky strain in the MD during the first part of the sequential biaxial deformation, UCW stretching in the MD and Hencky strain in the TD during the second part of the sequential biaxial deformation. Unlike the change in out-of-plane strain-optical behaviors for simultaneous biaxial stretching, almost linear strain-optical behaviors are observed throughout the experiment for all the film planes (MD-TD, TD-ND, MD-TD) at all stretching rates and ratios. The way the strain-optical behavior depends on the stretching rates is similar to dependencies uncovered in the simultaneous biaxial stretching. During the first part of the

166

sequential biaxial deformation, UCW in the MD, the in-plane (MD-TD) strain-optical behavior slightly increases while out-of-plane (TD-ND, MD-ND) strain-optical behavior more dramatically decreases. The opposite behaviors are observed during the second step of the sequential biaxial deformation. The in-plane (MD-TD) strain-optical behavior slightly decreases while out-of-plane (TD-ND, MD-ND) strain-optical behaviors dramatically increas. At lower stretching rates (i.e. 50 mm/min) the out-of-plane birefringence behaviors decrease more slowly than those of higher stretching rates (i.e 500 mm/min). As a result, by the end of the second step of the sequential biaxial stretching, the out-of-plane plane birefringence behaviors for the slower rates are higher; this is due to a higher initial level of birefringence at the beginning of the second step of the sequential biaxial deformation process. The DSC thermograph confirms that strain-induced crystallization takes place during the second stretching step. The total crystallinity at the end of sequential biaxial stretching at 50mm/min is about 5.4%. Furthermore, the DSC thermograph revealed that the amount of crystallinity at the early stage of UCW stretching in the second step is lower than that found at the end of the first step. This indicates that, during UCW stretching in the TD, crystals are broken up. The WAXD patterns reveal that the in-plane (MD-TD) is kept isotropic at all stretching rates and ratios while the out of planes (MD-ND, TD-ND) show some order in the polymer chain.

167

=5.4% =1.8%

=3.3% =1.67%

=1.57% =2.7% =1.3% =1.2%

FIGURE 6.10 STRAIN OPTICAL BEHAVIOR OF SEQUENTIAL BIAXIAL STRETCHED PI(BTDA-DAH) FILMS TO A DIFFERENT STRETCHING RATIOS AND RATES AT 90°C.

168

6.3.4 Structural Evolution

A schematic representation of structural evolution and a molecular model of

PI(BTDA-DAH) from an amorphous state to a final crystalline state during simultaneous and sequential biaxial stretching at 90°C at low stretching rates are provided in Figure 6.11 and 6.12 respectively. At the beginning of deformation, the polymer is in an amorphous state for both simultaneous and sequential biaxial stretching. Therefore, the polymer is isotropic in all of its planes. During simultaneous biaxial stretching, the polymer chains are stretched in both the MD and the TD. Therefore, the polymer is kept in an isotropic state in the MD-TD plane. Furthermore, this orientation is accompanied with a rotation of the phenyl groups, which start to become more parallel with the polymer specimen's surface.

Although, at higher stretching ratios, the polymer chains reach their full potential for extension, the polymer continues to be isotropic in the MD-TD plane. A higher level of phenyl groups parallel to the specimen surface appears. Moreover, strain-induced crystallization is promoted as the level of crystallinity increases.

During the first part of the sequential biaxial deformation, UCW stretching in the

MD, the polymer chains start to orient to the MD and strain-induced crystallization is promoted, which in turn tightens the network structure and results in a fast increase in the in-plane birefringence. This orientation forces the phenyl groups to become perpendicular to the PI(BTDA-DAH) specimen's surface. At early stretching ratios during the second step of sequential biaxial deformation, UCW stretching in the TD, some of the polymer chains start to orient more to the TD. Therefore, the MD-TD plane's anisotropy decreases; as a result, the in-plane birefringence decreases. This orientation forces the phenyl groups to

169

become parallel to the PI(BTDA-DAH) specimen's surface. Furthermore, the level of crystallinity decreases. The orientation in the TD results in breakage in the crystals structure. When the stretching ratios increase, the MD-TD become isotropic, and the phenyl groups become more parallel with the PI(BTDA-DAH) specimen's surface.

Furthermore, strain-induced crystallization is promoted as the level of crystallinity increases again.

170

FIGURE 6.11 SCHEMATIC REPRESENTATION OF THE STRUCTURAL EVOLUTION OF PI(BTDA-DAH) DURING SIMULTANEOUS BIAXIAL STRETCHING.

171

FIGURE 6.12 SCHEMATIC REPRESENTATION OF THE STRUCTURAL EVOLUTION OF PI(BTDA-DAH) DURING SEQUENTIAL BIAXIAL STRETCHING.

172

6.3.5 Dielectric properties.

Figure 6.13 shows the dielectric constant and loss of equal simultaneously biaxially stretched PI(BTDA-DAH) films which are stretched to stretching ratios of 1.4 X 1.4, 1.6

X 1.6 and 1.8 X 1.8. In this study, the effect of frequency, temperature, and stretching ratios on dielectric constant and loss are investigated. The dielectric constant of PI(BTDA-DAH) is sensitive to frequency, temperature and stretching ratios. First, the dielectric constant slightly decreases at higher frequencies. Second, the dielectric constant increases at higher temperatures. The mobility of the polar group influences the dielectric constant values.

When the polymer's free volume increases, the mobility of polymer chains also increases.

This allows the polar group to orient more freely, which increases the dielectric constant.

Third, the dielectric constant decreases with a decrease in stretching ratios. During the equal simultaneous biaxial stretching, the phenyl groups in the PI(BTDA-DAH) specimen's backbone become parallel to the specimen's surface, which reduce the polarizability along the specimen's thickness. Therefore, the dielectric constant decreases as the polarizability along the specimen's thickness decreases. Approximately 20% of the solvent is trapped in the PI(BTDA-DAH) solution cast film. As a result, the dielectric loss is higher than desirable (<1%). Figure 6.14 shows the dielectric loss of films which were vacuum compression pressed for 7 hours at 250°C after undergoing a solution-casting procedure; this was undertaken in order to evaporate the solvent in the polymer film. This method reduces the trapped solvent to around 0.1 %. As a result, the dielectric loss becomes less than 0.8% at frequencies between 1Hz and 1,000 Hz.

173

10 Hz. 100 Hz. 1000 Hz.

FIGURE 6.13 THE EFFECT OF FREQUENCY, TEMPERATURE AND STRETCHING RATIOS ON DIELECTRIC CONSTANT AND LOSS OF SIMULTANEOUS BIAXIAL STRETCHED PI(BTDA-DAH) FILMS.

174

FIGURE 6.14 THE EFFECT OF FREQUENCY, AND STRETCHING TYPES AND RATIOS ON DIELECTRIC LOSS OF BIAXIAL STRETCHED PI(BTDA-DAH) FILMS WHICH WERE COMPRESSION VACUUM MOLDED AFTER SOLUTION CASTING.

Figure 6.15 shows the dielectric constant at RT of simultaneously and sequentially biaxially stretched PI(BTDA-DAH) films which are stretched up to ratios of 1.6 X 1.6. In this study the films are vacuum-compressed for 7 hours at 250°C after a solution casting procedure. PI(BTDA-DAH) specimens which are UCW stretched to 1.6 X 1.6 (the first step of sequential biaxial stretching) have the highest dielectric constant. As discussed above, during UCW stretching, the polymer chains orient to the MD. This orientation forces the phenyl groups to become perpendicular to the PI(BTDA-DAH) specimen's surface, which increases the polarizability along the specimen's thickness. As a result, the dielectric constant increases. When the film is sequentially stretched to a higher level of equal biaxial stretching (the second step of sequential biaxial stretching), the polymer chains orient more to the TD and the alignment of the phenyl groups become more parallel

175

to the PI(BTDA-DAH) specimen's surface. As a result, the polarizability along the specimen's thickness decreases. Furthermore, the dielectric constant of films which are equal simultaneously biaxially stretched is higher than those which are equal sequentially biaxially stretched, due to a less parallel alignment of the phenyl groups with the PI(BTDA-

DAH) specimen's surface.

Sequential 1.6X1 Simultaneous 1.6X1.6

Sequential 1.6X1.6

Sequential As cast 1.6X1.3

FIGURE 6.15 THE EFFECT OF FREQUENCY, AND STRETCHING TYPES AND RATIOS ON DIELECTRIC CONSTANT OF BIAXIAL STRETCHED PI(BTDA-DAH) FILMS WHICH WERE COMPRESSION VACUUM MOLDED AFTER SOLUTION CASTING.

176

The breakdown strength of equal simultaneously biaxially stretched PI(BTDA-

DAH) films which were stretched to 1.4X1.4, 1.6X1.6 and 1.8X1.8 are described by the

Weibull plot shown in Figure 6.16. No significant effect on characteristic breakdown strength (휂 ), Weibull slope parameter (훽 ), and on the breakdown field strength are observed for the different stretching ratios.

FIGURE 6.16 THE EFFECT OF STRETCHING RATIOS ON DIELECTRIC BREAKDOWN OF SIMULTANEOUS BIAXIAL STRETCHED PI(BTDA-DAH) FILMS.

6.4 Conclusions

The deformation behavior of simultaneously and sequentially biaxially stretched

PI(BTDA-DAH) thin films in their rubbery state is studied. During simultaneous biaxial stretching, the polymer chains are oriented in both the MD and the TD, and the refractive indices in the MD and the TD stay almost the same. As a result, the isotropy in the MD-

177

TD remains throughout the experiment. This orientation is accompanied with the preferential alignment of the phenyl groups as parallel to the MD-TD surface. Therefore, the refractive index in the ND decreases with higher stretching ratios. As a result, polymer polarizability along the specimen's thickness decreases, which reduces the total dielectric constant along the specimen's thickness. During the first step of sequential biaxial stretching, the polymer chains are oriented in the MD, which is accompanied with the preferential alignment of the phenyl groups as perpendicular to the MD-TD surface.

Therefore, the refractive indices in the MD and the ND increase while the refractive index in the TD decreases. As a result, the polymer's polarizability is increased along its thickness. During the second step of sequential biaxial stretching, the polymer chains are more oriented in the TD, which is now accompanied by the preferential alignment of the phenyl groups as parallel to the MD-TD surface. Therefore, the refractive indices in the

MD and the ND decrease while the refractive index in the TD increases. As a result, the polymer's polarizability is decreased along its thickness. The dielectric constant is also found to be sensitive to the temperature and frequency of stretching. The dielectric loss is found to be sensitive to the amount of trapped solvent in the polymer films. A higher concentration of trapped solvent results in a higher dielectric loss.

178

6.5 Appendix

TABLE 6.1 REFRACTIVE INDICES OF PI(BTDA-DAH) WHICH ARE STRETCHED TO DIFFERENT STRETCHING RATIOS DURING SIMULTANEOUS AND SEQUENTIAL BIAXIAL STRETCHING AT 90°C AND 50 MM/MIN.

Stretching ratio Stretching type 푛푀퐷 푛푇퐷 푛푁퐷 휆푀퐷×휆푇퐷 1.0X1.0 As cast 1.6325 1.6337 1.6307 1.2X1.2 1.6304 1.6317 1.6310 1.4X1.4 1.6343 1.6337 1.6154 simultaneous 1.6X1.6 1.6350 1.6356 1.6137 1.8X1.8 1.6363 1.6382 1.6066 1.2X1.0 1.6310 1.6330 1.6458 1.2X1.2 1.6337 1.6343 1.6237 1.4X1.0 1.6337 1.6304 1.6822 1.4X1.2 1.6375 1.6330 1.6407 1.4X1.4 1.6327 1.6356 1.6161 1.6X1.0 1.6379 1.6297 1.7191 1.6X1.2 1.6363 1.6310 1.6597 sequential 1.6X1.4 1.6337 1.6343 1.6055 1.6X1.6 1.6330 1.6388 1.5764 1.8X1.0 1.6379 1.6290 1.7194 1.8X1.2 1.6356 1.6297 1.6898 1.8X1.4 1.6343 1.6330 1.6270 1.8X1.6 1.6317 1.6343 1.5990 1.8X1.8 1.6323 1.6350 1.5410

179

CHAPTER VII

CONCLUSION

In the present study investigation on the behavior of novel polymers thin films for capacitors application under different types of processing: uniaxial, biaxial, and relaxation were performed using novel instruments which allow real-time measurement of the mechano-optical and IR dichroism behavior. As a result, true stress - Hencky strain, true stress - birefringence, Hencky strain – birefringence, true stress – orientation function, and

Hencky strain – orientation function measured. Additional off-line measuring techniques such as, WAXD and differential scanning calorimetry, where used to assess the structural evaluation of these novel polymers during the processing procedure. An Abbe refractometer, an IMASS time domain dielectric spectrometer, a Sawyer-Tower circuit, and a resistor capacitor (RC) circuit were used to investigate the effect of the processing conditions and the structure evaluation.

The first study focuses of the investigation of the structural evaluation of PP and

PPOH during processing cycle of heating, uniaxial stretching, annealing and cooling. In this processing cycle, the polymers are subjected to complex thermo-mechanical deformation. The real time in-plane birefringence and orientation functions of the amorphous and crystalline segments assess to reveal the morphology of crystal structures

180

during the processing cycle. At early stages of uniaxial deformation of PP, the spherulitic break-up led to break up and rotation of lamellae fragments in the equatorial regions of the spherulites. Further stretching reversed this trend. The amorphous and crystalline chains attained increasingly large preferential orientation in the MD. The rotation of lamellae fragments in the equatorial regions of the spherulites was not observed for PPOH.

The second study focuses of the investigation of the structural evaluation of

PI(BTDA-DAH) during uniaxial deformation. The effect of processing conditions, temperatures, and stretching on the mechno-optical behavior of PI(BTDA-DAH) thin films were studied. At stretching temperatures above Tll and low stretching rates, three regimes of stress optical behavior were observed. The first regime was linear and followed the stress optical rule; the stress optical constant was found to be 17.8 GPa-1. In this regime the polymer structure remained amorphous. In regime II, a rapid increase in birefringence with a modest increase in stress was observed; it was caused by strain-induced crystallization.

In regime IIIc the polymer chains reached their extension limitation; therefore, the birefringence leveled off with a high increase in stress. The PI(BTDA-DAH) did not develop a three-dimensional crystalline order during the uniaxial deformation.

Furthermore, irregularity in the network structure order was observed. At stretching temperatures below Tll the stress optical behavior follows regimes I-IIIa. In regime I, the birefringence linearly increases with stress. This is followed by a negative deviation from linearity, which slowly increases in birefringence until a plateau is reached (regime IIIa), which polymer chains reached their extension limitation. In this case the strain-induced crystallization does not occur.

181

The third study focuses of the investigation of the structural evaluation of

PI(BTDA-DAH) during uniaxial deformation follows by relaxation. The effect of processing conditions: deformation ratio and relaxation time on the mechno-optical- orientation function behaviors of PI(BTDA-DAH) thin films were studied. The mechno- optical and mechano-orientation function behaviors found to be similar to each other.

Furthermore, during relaxation they are found to depend on the amount of deformation undergone during uniaxial stretching. Three regimes of these behavior are revealed during the relaxation. In regime I, during the early stage of deformation, birefringence and orientation function linearly decrease with stress during relaxation. In regime II, the birefringence and orientation function behaviors depend on the level of the stress deformation undertaken during uniaxial stretching. At the early stage of deformation in regime II, birefringence and orientation function decrease with the stress during relaxation.

At the early intermediate stage of deformation, the birefringence and orientation function continue to decrease with the stress during relaxation until a critical point at which birefringence and orientation function start to increase. At the intermediate stage of deformation in regime II, the stress relaxation is accompanied by no change in birefringence until a critical point at which the birefringence starts to increase. At the high stage of deformation in regime II, the stress relaxation is accompanied by increasing birefringence and orientation function. In regime III, an increase in birefringence and orientation function are observed. The increase in birefringence and orientation function is a result of a spontaneous deformation process which occurs during relaxation. The level of crystallinity and the perfection of the crystals increases as a results of an increase in the relaxation time allotted for specimens which are stretched in regimes II and III. Moreover,

182

the relaxation time found has a critical role on the breakdown strength. Higher relaxation time improves the maximum breakdown strength of the PI(BTDA-DAH) films.

The last study focuses of the investigation of the structural evaluation of PI(BTDA-

DAH) during simultaneously and sequentially biaxial deformation. The orientation of the polymer chain during both of the biaxial methods is accompanied with the preferential alignment of the phenyl groups as parallel to the MD-TD surface. Therefore, the refractive index in the ND decreases. As a result, polymer polarizability along the specimen's thickness decreases, which reduces the total dielectric constant along the specimen's thickness. The level of the alignment of the phenyl groups as parallel to the MD-TD surface is found to be higher for sequentially biaxial deformation compare simultaneously biaxial deformation for the same stretching ratios and stretching rate. As a result, the dielectric constant along the specimen's thickness for PI(BTDA-DAH) film which was stretched by simultaneously biaxial deformation method is higher than those which was stretched by sequentially biaxial deformation method. In contrast, stretching PI(BTDA-

DAH) film in UCW aligned the phenyl groups as perpendicular to the MD-TD surface. As a result of this conformation, the polymer's polarizability is increased along its thickness.

Therefore, dielectric constant along the specimen's thickness increases.

183

REFERENCES

1. Kalhammer , F. R. Energy-Storage Systems. Sci. Am.; (United States) 1979, 241 (6), 56-65.

2. Ribeiro, P. F.; Johnson, B. K.; Crow, M. L.; Arsoy, A.; Liu, Y. Energy Storage Systems for Advanced Power Applications. PROCEEDINGS OF THE IEEE 2011, 89 (12), 1744-1756.

3. Chen, H.; Cong, T. N.; Yang, W.; Tan, C.; Li, Y.; Ding, Y. Progress in electrical energy storage system: A critical review. Progress in Natural Science 2009, 19, 291- 312.

4. Omer, A. M. Energy, environment and sustainable development. Renewable and Sustainable Energy Reviews 2008, 12, 2265–2300.

5. Gan, L.; Eskeland, G. ; Kolshus, H. H. Green electricity market development: Lessons from Europe and the US. Energy Policy 2007, 35, 144-155.

6. Ackermann, T.; Andersson, G.; SoÈder, L. Overview of government and market driven programs for the promotion of renewable power generation. Renewable Energy 2011, 22, 197-204.

7. Electropaedia, http://www.mpoweruk.com/alternatives.htm (accessed August 8, 2016).

8. Khaligh, A.; Li, Z. Battery, Ultracapacitor, Fuel Cell, and Hybrid Energy Storage Systems for Electric, Hybrid Electric, Fuel Cell, and Plug-In Hybrid Electric

184

Vehicles: State of the Art. IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY 2010, 59 (6), 2806-2814.

9. Ibrahim, H.; Ilinca, A.; Perron, J. Energy storage systems—Characteristics and comparisons. Renewable and Sustainable Energy Reviews 2008, 12, 1221-1250.

10. Chung , T. C. M. Functionalization of Polypropylene with High Dielectric. Green and Sustainable Chemistry 2012, 2, 29-37.

11. Kaiser, J. C. Fundamental for all capacitors. In The Capacitor Handbook; Kaiser, J. C. Van Nostrand Reinhold: New Youk, 1993; pp 1.

12. Maxwell, J. C. A Dynamical Theory of the Electromagnetic Field. Ohilosophical Transactions 1865, 155, 459-512.

13. Abramowitz, M.; Parry-Hill, M. J.; Davidson, M. W. Electromagnetic Radiation. https://micro.magnet.fsu.edu/primer/java/scienceopticsu/electromagnetic/index.html (accessed July 12, 2016).

14. Wikipedia. https://en.wikipedia.org/wiki/Ray_(optics) (accessed August 5, 2016).

15. Murphy, B.; Spring, K. R.; Fellers, T. J.; Davidson, M. W. Nikon. http://nikon.magnet.fsu.edu/print/articles/polarized/birefringence-print.html (accessed August 5, 2016).

16. Samuels, R. J. Application of refractive index measurements to polymer analysis. Journal of Applied Polymer Science 1981, 26 (4), 1383-1412.

17. Cakmak , M.; White, J. L.; Spruiell, J. E. Structural Characterization of Crystallinity and Crystalline Orientation in Simultaneously Biaxially Stretched and Annealed

185

Polyethylene Terephthalate Films. Journal of Polymer Engineering 1986, 6 (1-4), 291-312.

18. Stein, R. S.; Onogl, S.; Keedy, D. A. The dynamic birefringence of high polymers. Journal of Polymer Science 1962, 57, 801-821.

19. Stein, R. S.; Onogi, S.; Sasaguri, K.; Keedy, D. A. Dynamic Birefringence of High Polymers II. Journal of Applied Physics 1963, 34, 80-89.

20. Galay, L.; Cakmak, M. Online Monitoring of Birefringence Developmentin Heat- Setting Polymer Films with a Fast Dual-Wavelength Optical Technique. I. Uniaxially Oriented Poly(ethylenenaphthalate). Journal of Polymer Science: Part B: Polymer Physics, 2001, 39, 1107-1121.

21. Yang, H. H.; Chouinard, M. P.; Lingg, E. I. Birefringence of Highly Oriented Fibers. Journal of Polymer Science: Polymer Physics Edition 1982, 20, 981-987.

22. Redner , A. S. Photoelastic Measurements by Means of Computer-Assisted Spectral- Contents Analysis. ExperimentaI Mechanics 1985, 25 (2), 148-153.

23. Ryu, D. S.; Inoue, T.; Osaki, K. A birefringence study of polymer crystallization in the process of elongation of films. Polymer 1998, 12, 2515-2520.

24. Matsumoto, T.; Bogue, D. C. Stress Birefringence in Amorphous Polymers under Nonisothermal Conditions. Journal of Polymer Science: Polymer Physics Edition 1977, 15, 1663-1674.

25. Hiemenz, C. P.; Lodge, P. T. Polymer Chemistry, 2nd ed.; CRC Press: Boca Raton, 2007.

186

26. Cakmak, M.; Simhambhatla, M. Dynamics od Uni and Biaxial Deformation and Its Effects on the Thickness Uniformity and Surface Rouhness of Poly(Ether Ether Ketone) Films. Polymer Engineering & Science 1995, 19, 1562-1568.

27. Mulligan, J.; Cakmak, M. Nonlinear Mechanooptical Behavior of Uniaxially Stretched Poly(lactic acid): Dynamic Phase Behavior. Macromolecules 2005, 38, 2333-2344.

28. Martins, C. I.; Cakmak, M. Large Deformation Mechano-Optical and Dynamical Phase Behavior in Uniaxially Stretched Poly(ethylene naphthalate). Macromolecules 2005, 38, 4260-4273.

29. Kanuga, K.; Cakmak, M. Nonlinear Mechanooptical Behavior of Poly(ethylene naphthalate)/Poly(ether imide) Blends. Dynamic Phase Behavior. Macromolecules 2005, 38, 9698–9710.

30. Rao, I. J.; Rajagopal, K. R. A study of strain-induced crystallization of polymers. International Journal of Solids and Structures 2001, 38, 1149-1167.

31. Janeschitz-Kriegl, H.; Ratajski, E.; Wippel, H. The physics of athermal nuclei in polymer crystallization. Colloid and Polymer Science 1999, 27 (2), 217-226.

32. Magill, J. H. Crystallization of Poly‐(Tetramethyl‐p‐Silphenylene)‐Siloxane Polymers. Journal of Applied Physics 1964, 35 , 3249-3259.

33. Chien, C. M.; Weiss, A. R. Strain-Induced Crystallization Behavior of Poly(Ether Ether Ketone) (PEEK). Polymer Engineering And Science 1988, 28 (1), 6-12.

34. Bourvellect, l. G.; Monnerie, L.; Jarry, J. P. Kinetics of induced crystallization during stretching and annealing of poly(ethylene terephthalate) films. Polymer 1987, 28, 1712-1716.

187

35. IHS Markit. https://www.ihs.com/products/polypropylene-resins-chemical- economics-handbook.html (accessed Aug. 14, 2016).

36. NATPET. http://www.natpet.com/natpet-media-center/polypropylene-the-material- of-choice/ (accessed Aug. 14, 2016).

37. Galeski, A. Nucleation of polypropylene. In Polypropylene Structure, blends and composites: Volume 1 Structure and Morphology, 1st ed.; Karger-Kocsis, J., Ed.; Chapman & Hall: Cambridge, 1995; Vol. 1, pp 116-123.

38. Maier, C.; Calafut, T. Polypropylene - The Definitive User's Guide and Databook; lastics Design Library a division of William Andrew Inc: Norwich, 1998.

39. Takekoshi, T. Synthesis of Polyimides. In Polyimides: Fundamentals and Applications; Ghosh, M. K., Mittal, K. L., Eds.; Marcel Dekker, Inc.: New York, 1996; p 7.

40. Ree, M. High performance polyimides for applications in microelectronics and flat panel displays. Macromolecular Research 2006, 14, 1-33.

41. Nishino, T.; Kotera, M.; Inayoshi, N.; Miki, N.; Nakamae, K. Residual stress and microstructures of aromatic polyimide with different imidization processes. Polymer 2000, 41, 6913-6918.

42. Meador, M. A. B.; Malow, E. J.; Silva, R.; Wright, S.; Quade, D.; Vivod, S. L.; Guo, H.; Guo, J.; Cakmak, M. Mechanically Strong, Flexible Polyimide Aerogels Cross- Linked with Aromatic Triamine. Applied Materials & Interfaces 2012, 4 (2), 539- 544.

188

43. Southward, R. E.; Thompson, D. W. Reflective and conductive silvered polyimide films for space applications prepared via a novel single-stage self-metallization technique. Materials and Design 2001, 22, 565-576.

44. Yu , D.; Yu, L. Design and Synthesis of Functionalized Polyimides for Second-Order Nonlinear Optics. Macromolecules 1994, 27, 6718-6721.

45. Harasimowicz , H.; Orluk, P.; Zakrzewska-Trznadel, G.; Chmielewski, A. G. Application of polyimide membranes for biogas purification and enrichment. Journal of Hazardous Materials 2007, 144, 698-702.

46. Sroog, C. E. History of the Invention and Development of the Polyimides. In Polyimides: Fundamentals and Applications; Ghosh , M. K., Mittal, K. L., Eds.; Marcel Dekker: New York, 1996; pp 1-6.

47. Garcı´a, M. G.; Marchese, J.; Ochoa, N. A. Aliphaticearomatic polyimide blends for H2 separation. International Journal of Hydrogen Energy 2010, 35, 8983-8992.

48. Takekoshi, T. Synthesis of Polyimides. In Polyimides: Fundamentals and Applications; Ghosh, M. K., Mittal, K. L., Eds.; Marcel Dekker: New York, 1996; pp 7-48.

49. Kudryavtsev, V. V. Polyimide Synthesis by Chemical Cyclization of Polyamic Acids. In Polyamic Acids and Polyimides: Synthesis, Transformations, and Structure; Bessonov, M. I., Zubkov, V. A., Eds.; CRC Press, Inc.: Boca Raton, 1993; pp 1-46.

50. Diaham, S.; Locatelli, M. L.; Khazaka, R. BPDA-PDA Polyimide: Synthesis, Characterizations, Aging and Semiconductor Device Passivation, 2016. IN TECH. ations, Prof. Marc Abadie (Ed.), InTech, (accessed May 22, 2015).

189

51. Breitenbach, J. Melt extrusion: from process to drug delivery technology. European Journal of Pharmaceutics and Biopharmaceutics 2002, 54, 107-117.

52. Robertson, G. L. Food Packaging Principles and Practice, 3rd ed.; CRC press: Boca Raton, 2012.

53. Xu, C.; Boggs, S. Measurement of Resistive and Absorption Currents in Capacitor Films up to Breakdown. IEEE International Symposium on Electrical Insulation, Toronto, 2006; pp 249–252.

54. Wagner Jr., J. R. Multilayer Flexible Packaging: Technology and Applications for the Food, Personal Care, and Over-the-Counter Pharmaceutical Industries; Elsevier: Burlington, 2010.

55. Demeuse, M. T. Fundamentals of biaxial stretching and definitions of terms. In Biaxial Stretching of Film - Principles and Applications; Demeuse, M. T., Ed.; Woodhead Publishing Limited: Cornwal, 2011; pp 1-13.

56. McKeen , L. W. Production of films, containers, and membranes In Permeability Properties of Plastics and Elastomers: A Guide to Packaging and Barrier Materials, 3rd ed.; Elsevier: Waltham, 2012.

57. Brink T. The Extrusionists- Films. http://extrusionist.com/extrusion- processes/films/index.html (accessed August 16, 2016).

58. Koike, Y.; Cakmak, M. Real time development of structure in partially molten state stretching of PP as detected by spectral birefringence technique. polymer 2003, 44, 4249-4260.

59. Samules, R. J. Quantitative characterization of deformation in polypropylene fibers. ournal of Polymer Science Part A‐2: Polymer Physics 1968, 6, 2021-2041.

190

60. Okajima, S.; Kurihara , K.; Homma, K. Studies on the Stretching of Polypropylene Film. I. Two-step Biaxial Stretching. Journal of Applied Polymer Science 1967, 11, 1703-1717.

61. Okajima, S.; Homma, K. Studies on the Stretching of Polypropylene Film. II. Polyaxial Stretching. Journal of Applied Polymer Science 1968, 12, 411-423.

62. Tanaka, H.; Masuko, T.; Homma, K.; Okajima, S. Studies on the Biaxial Stretching of Polypropylene Film. III. Electron Microscopic Observation of the One-Step Biaxial Stretching of Isotactic Polypropylene Spherulites. Journal of Applied Polymer Science: Part A-1 1969, 7, 1997-2006.

63. Masuko, T.; Tanaka, H.; Okajima, S. Studies on Biaxial Stretching of Polypropylene Film. IV. Overall Orientation during One-Step Biaxial Stretching and its Application to the Estimation of the Principal Refractive Indices of Isotactic Polypropylene. Journal of Applied Polymer Science: Part A-2 1970, 8, 1565-1574.

64. Okajima, S.; Iwato, N.; Tanaka, H. Studies on Biaxial Stretching of Polypropylene film. VII. Conditions od Two-Step Biaxial Stretching and Type of Crystaaline Orientation. Polymer letters 1971, 9, 797-804.

65. Tanaka, H.; Masuko, T.; Okajima, S. Studies on the Biaxial Stretching of Polypropylene Film. VII. Refractive Index of Film Measured with Abbe Refractometer. Journal of Applied Polymer Science 1972, 16, 441-447.

66. Iwato, N.; Tanaka, H.; Okajima, S. Studies on Biaxial Stretching of Polypropylene Film. X. High-Temperature X-Ray and Optical Study of Type I11 Orientation. Journal of Applied Polymer Science 1973, 17, 2533-2546.

67. Tanaka, H.; Okajima, S. Studies on The Biaxial Stretching of Polypropylene film. XII. Effect of Crystallinity on The Deformation Mechanism of The Biaxial Stretching in One Step. Polymer letters edition 1977, 15, 349-362.

191

68. Griffiths, P. R.; de Haseth, J. A.; Fourier Transform Infrared Spectrometry, Second Edition; John Wiley & Sons, Inc: Ney Jersey, 2007; Vol. 83.

69. Fraser, R. D. B. Interpretation of Infrared Dichroism in Axially Oriented Polymers. The Journal of Chemical Physics 1958, 28, 1113-1115.

70. Yeh, F.; Hsiao, B. S.; Sauer, B. B.; Michel, S.; Siesler, H. W. In-Situ Studies of Structure Development during Deformation of a Segmented Poly(urethane-urea) Elastomer. Macromolecules 2003, 36, 1940-1954.

71. Shigematsu, Y.; Takada, A.; Nemoto, N.; Nitta, K.-H. An instrument for simultaneous kinetic measurements of microscopic infrared dichroism and stress of inhomogeneous polymer thin films at constant elongation rate. Review of Scientific Instruments 2001, 72 (10), 3927-3928.

72. Manning, C. J. Tilt-compensated interferometers. US 6967722 B2, November 22, 2005.

73. Unsal, E.; Cakmak, M. Molecular mechanism of temporal physico/chemical changes that take place during imidization of polyamic acid: Coupled real-time rheo-optical and IR dichroism measurements. Polymer 2014, 55, 6569-6579.

74. Offenbach, I.; Gupta, S.; Ma, R.; Sotzing, G. A.; Weiss, R. A.; Cakmak, M. REAL- TIME Mechano-Optical Behavior And Structural Evolution Of Polyimide (Btda- Dah) During Uniaxial Deformation. to be submitted for publication, 2016.

75. Samuels, R. J. Infrared dichroism, molecular structure, and deformation mechanisms of isotactic polypropylene. Die Makromolekulare Chemie Supplement 1981, 3, 241- 270.

192

76. Bao, R.; Ding, Z.; Zhong, G.; Yang, W.; Xie, B.; Yang, M. Deformation-induced morphology evolution during uniaxial stretching of isotactic polypropylene: effect of temperature. Colloid and Polymer Science 2012, 290, 261-274.

77. Hasegawa, M.; Okuda, K.; Horimoto, M.; Shindo, Y.; Yokota, R.; Kochi, M. Spontaneous Molecular Orientation of Polyimides Induced by Thermal Imidization. 3. Component Chain Orientation in Binary Polyimide Blends. Macromolecules 1997, 30, 5745-5752.

78. Barshaw, E. J.; White, J.; Chait, M. J.; Cornette, J. B.; Bustamante, J.; Folli, F.; Biltchick, D.; Borelli, G.; Picci, G.; Rabuffi, M. High Energy Density (HED) Biaxially-Oriented Poly-Propylene (BOPP) Capacitors For Pulse Power Applications. IEEE TRANSACTIONS ON MAGNETICS 2007, 43 (1), 223-225.

79. Rabuffi, M.; Picci, G. Status Quo and Future Prospects for Metallized Polypropylene Energy Storage Capacitors. IEEE TRANSACTIONS ON PLASMA SCIENCE 2002, 30 (5), 1939-1942.

80. Wang, C. C.; Ramprasad, R. Dielectric properties of organosilicons from first principles. Journal of Materials Science 2011, 90-93, 46.

81. Yuan, X.; Matsuyama, Y.; Chung, T. C. M. Synthesis of Functionalized Isotactic Polypropylene. Macromolecules 2010, 43, 4011-4015.

82. Lorenzini, R. G.; Kline, W. M.; Wang, C. C.; Ramprasad, R.; Sotzing, G. A. The rational design of polyurea & polyurethane dielectric materials. Polymer 2013, 54, 3529-3533.

83. Gupta, S.; Yuan, X.; Chung, M. T. C.; Kumar, S.; Cakmak, M.; Weiss, R. A. Effect of Hydroxyl-Functionalization on the Structure and Properties of Polypropylene. Macromolecules 2013, 46 (14), 5455-5463.

193

84. Gupta, S.; Yuan, X.; Chung, T. C. M.; Cakmak, M.; Weiss, R. A. Isothermal and nonisothermal crystallization kinetics of hydroxyl-functionalized polypropylene. Polymer 2014, 55, 924-935.

85. Henze, E. D.; Kahn, H. A. Process for production of biaxially oriented polypropylene film. US3223764 A, Dec 14, 1965.

86. Jolliffe, C. N. Method of making biaxially oriented thermoplastic films. US5429785 A, July 4, 1995.

87. Cole, K. C.; Depecker, C.; Jutigny, M.; Lefebvre, J. M.; Krawcza, P. Biaxial Deformation of Polyamide-6: Assessment of Orientation by Means of Infrared Trichroism. Polymer Engineering & Science 2004, 44 (2), 231-240.

88. Koike, Y.; Cakmak, M. Atomic Force Microscopy Observations on the Structure Development during Uniaxial Stretching of PP from Partially Molten State: Effect of Isotacticity. Macromolecules 2004, 37, 2171-2181.

89. Unsal, E.; Nugay, I. I.; Offenbach, I.; Gross, M.; Manning, C.; Cakmak, M. Rapid integrated rheo-optical and polarized Fourier-transform infrared spectrometry measurement system for polymer films undergoing chemomechanical changes. Review of Scientific Instruments 2013, 84, 073901.

90. Wunderlich, B. Macromolecular Physics; Academic Press: New, 1980; Vol. 3, p 63.

91. Theory of crystallization in copolymers. Transactions of the Faraday Society 1955, 51, 848-857.

92. Gupta, S.; Yuan, X.; Chung, T. C. M.; Cakmak, M.; Weiss, R. A. Unpublished results.

194

93. McDonald, M. P.; Ward, I. M. The assignment of the infra-red absorption bands and the measurement of tacticity in polypropylene. Polymer 1961, 2, 341-355.

94. Fetters, L. J.; Lohse, D. J.; Richter, D.; Witten, D. A.; Zirkel, A. Connection between Polymer Molecular Weight, Density, Chain Dimensions, and Melt Viscoelastic Properties. Macromolecules 1994, 27 (17), 4639-4647.

95. Sasaguri, K.; Hoshino, S.; Stein, R. S. Relationship between Morphology and Deformation Mechanisms of Polyolefins. Journal of Applied Physics 1964, 35, 47- 54.

96. Song, Y.; Nitta, K.-h.; Nemoto, N. Deformation Mechanisms of Polymer Thin Films by Simultaneous Kinetic Measurement of Microscopic Infrared Dichroism and Macroscopic Stress. 2. Molecular Orientation during Necking Process of Isotactic Polypropylene. Macromolecules 2003, 36, 1955-1961.

97. Bao, R.; Ding, Z.; Zhong, G.; Yang, W.; Xie, B.; Yang, M. Deformation-induced morphology evolution during uniaxial stretching of isotactic polypropylene: effect of temperature. Colloid and Polymer Science 2012, 290 (3), 261-274.

98. Shimamura, K.; Murakami , S.; Katayama, K.-i. Internal morphology of uniaxially drawn polyethylene spherulites. Die Makromolekulare Chemie, Rapid Communications 1982, 3 (4), 199-204.

99. Michler, G. H. Correlations between molecular weight, morphology and micromechanical deformation processes of polyethylenes. Colloid Polymer Science 1992, 270 (7), 627-638.

100. Sasaguri, K.; Yamada, R.; Stein, R. S. J. Relationship between Morphology and Deformation Mechanisms of Polyolefins. II. Journal of Applied Physics 1964, 35, 3188-3194.

195

101. Drechsler, D.; Karbach, A.; Fuchs, H. Morphology and In Situ Deformation of Polyamide Films Investigated with Scanning Force Microscopy. Surface and Interface 1997 , 25, 537–542.

102. Onogi, S.; Asada, T.; Tanaka, A. Rheo-optical studies of high polymers. XIV. Study of the deformation mechanism in polymer blends of polypropylene with ethylene– propylene rubber. Journal of Polymer Science Part A-2: Polymer Physics 1969, 7, 171-182.

103. Muller, R.; Pesce, J. J. Stress-optical behaviour near the Tg and melt flow-induced anisotropy in amorphous polymers. Polymer 1994, 35, 734–739.

104. RabieJ, S.; Wlochowicz, A. SAXS and WAXS investigations of the crystallinity in polymers. Die Angewandte Makromolekulare Chemie 1990, 175 (1), 81-97.

105. Agarwal, P. K.; Somani, R. H.; Weng, W.; Mehta, A.; Yang, L.; Ran, S.; Liu, L.; Hsiao, B. S. Shear-Induced Crystallization in Novel Long Chain Branched Polypropylenes by in Situ Rheo-SAXS and -WAXD. Macromolecules 2003, 36, 5226-5235.

106. Samuels, R. J. Morphology of deformed polypropylene. Quantitative relations by combined x-ray, optical, and sonic methods. Journal of Polymer Science Part A 1965, 3, 1741-1763.

107. Sharma, P.; Bhatti, T. S. A review on electrochemical double-layer capacitors. Energy Conversion and Management 2010, 51, 2901-2912.

108. Barker, P. P. Ultracapacitors for Use in Power Quality and Distributed Resource Applications. Power Engineering Society Summer Meeting, 2002 IEEE, Chicago, 2002; pp 316-320.

196

109. Drolia, A.; Jose , P.; Mohan , N. An approach to connect ultracapacitor to fuel cell powered electric vehicle and emulating fuel cell electrical characteristics using switched mode converter. Industrial Electronics Society, 2003. IECON '03. The 29th Annual Conference of the IEEE, Roanoke, 2003; pp 897-901.

110. Pagano, M.; Piegari, L. Hybrid Electrochemical Power Sources for Onboard Applications. IEEE Transactions on Energy C 2007, 22, 450-456.

111. Slenes, K. M.; Winsor, P.; Scholz, T.; Hudis, M. Pulse Power Capability of High Energy Density Capacitors Based on a New Dielectric Material. IEEE Transactions on Magnetics 2001, 37 (1), 324-327.

112. Beach, F. C.; McNab, I. R. PRESENT AND FUTURE NAVAL APPLICATIONS FOR PULSED POWER; Naval Directed Energy & Electric Weapons Naval Sea: Washington D.C, 2005.

113. Ho, J.; Jow, R. Characterization of High Temperature Polymer Thin Films for Power Conditioning Capacitors; Final; Army Research Laboratory: Adelphi, 2009.

114. Irwin, .; Cao, Y.; Tan, Q.; Younsi,. High temperature capacitors and method of manufacturing the same. US20070258190 A1, Nov. 8, 2007.

115. Saia, R. ; Durocher, K. ; Gorowitz, B. Fabrication method for thin film capacitors. US5736448 A, Apr 7, 1998.

116. Muruganand, S.; Narayandass, S. K.; Mangalaraj, D.; Vijayan, T. Dielectric and conduction properties of pure polyimide films. Polymer International 2001, 50, 1089-1094.

117. Chahal, C.; Tummala, R. R.; Allen, M. G.; Swaminathan, M. A Novel Integrated Decoupling Capacitor for MCM-L Technology. IEEE Transactions on Components, Packaging, and Manufacturing Technology- Part B 1998, 21 (2), 184-193.

197

118. Devaraju, N. G.; Kim, E. S.; Lee, B. I. The synthesis and dielectric study of BaTiO3/polyimide nanocomposite films. Microelectronic Engineering 2005, 82, 71- 83.

119. Ma, R.; Baldwin, A. F.; Wang, C.; Offenbach, I.; Cakmak, M.; Ramprasad, R.; Sotzing, G. A. Rationally Designed Polyimides for High-Energy Density Capacitor Applications. ACS Applied Materials & Interfaces 2014, 6, 10445-10451.

120. Kolesov, S. N. IEEE Transactions on Electrical Insulation 1980 1980, EI-15 (5), 382-388.

121. BOWDEN, P. B.; YOUNG, R. J. Deformation mechanisms in crystalline polymers. Journal of Materials Science 1974, 9, 2034-2051.

122. Guan, F.; Pan, J.; Wang, J.; Wang, Q.; Zhu, L. Crystal Orientation Effect on Electric Energy Storage in Poly(vinylidene fluoride-co-hexafluoropropylene) Copolymers. Macromolecules 2010, 43, 384-392.

123. Offenbach, I.; Gupta, S.; Treich, G.; Sotzing, G. A.; Weiss, R. A.; Cakmak, M. Role of relaxation on strain induced crystallization of uniaxially stretched PI(BTDA- HDA) films. To be submitted for publication 2016.

124. Kokturk, G.; Serhatkulu, T. F.; Cakmak, M.; Piskin, E. Evolution of Phase Behavior and Orientation in Uniaxially Deformed Polylactic Acid Films. Polymer Engineering & Science 2002, 42, 1619-1628.

125. Sen, T. Z.; Toki, S.; Cakmak, M. Real time development of stress and birefringence in PET during uniaxial stretching as detected by spectral birefringence technique. Annual Technical Conference - Society of Plastics Engineers , 2001; pp 1510-1514.

198

126. SONG, S. S.; WHITE, J. L.; CAKMAK, M. Crystallization Kinetics and Nucleating Agents for Enhancing the Crystallization of Poly(p-Phenylene Sulfide). Polymer Engineering & Science 1990, 30, 944−949.

127. Seif, S.; Cakmak, M. Stress - Optical behavior of Poly(m-xylylenediamine adipamide) (Nylon MXD6): Influence of molecular weight. Polymer 2010, 51, 3762- 3773.

128. BOYER, R. F.; HEESCHEN, J. P.; GILLHAM, J. K. Comments on a Paper by Axelson and Mandelkern Concerning I3C NMR Spectra of Polymers. Journal of Polymer Science: Polymer Physics Edition 1981, 19, 13-21.

129. BOYER, R. F. Contributions of Torsional Braid Analysis to Tll. Polymer Engineering & Science 1979, 19, 732-748.

130. LOBANOV, A. M.; FRENKEL, S. Y. THE NATURE OF THE S0-CALLED "LIQUID-LIQUID" TRANSITION IN POLYMER MELTS. A REVIEW. Polymer Science U.S.S.R 1980, 22 (5), 1150-1163.

131. PLAZEK, D. J. Some Evidence Against the Existence of the Liquid-Liquid Transition. I. Transitions or Dispersions: A Quantitative Explanation of One Torsional Braid Log Decrement Peak. Journal of Polymer Science: Polymer Physics Edition 1982, 20, 1533-1550.

132. Kisliuk, A.; Mathers, R. T.; Sokolov, A. P. Crossover in Dynamics of Polymeric Liquids: Back to Tll? Journal of Polymer Science: Part B: Polymer Physics 2000, 38, 2785-2790.

133. Murthy, S. S. N. Liquid-Liquid Transition in Polymers and Class-Forming Liquids. Journal of Polymer Science: Part B: Polymer Physics 1993, 31, 475-480.

199

134. Suehiro, K.; Chatani, Y.; Tadokoro, H. Structural Studies of Polyesters. VI. Disordered Crystal Structure (Form II) of Poly(β-propiolactone). Polymer Journal 1975, 7, 352-358.

135. Tadokoro, H. Strucure of crystalline polymers; Johan Wiley & Sons: New York, 1976; pp 146-159.

136. Chang, S.; Han, C. D. Effect of Flexible Spacer Length on the Phase Transitions and Mesophase Structures of Main-Chain Thermotropic Liquid-Crystalline Polyesters Having Bulky Pendent Side Groups. Macromolecules 1997, 30, 1670-1684.

137. Atkins, E. D. T.; Hill, M.; Hong, S. K.; Keller, A.; Organ, S. Lamellar Structure and Morphology of Nylon 46 Crystals. A New Chain Folding Mechanism for Nylons. Macromolecules 1992, 25, 917-924.

138. McNab, I. R. Launch to Space With an Electromagnetic Railgun. IEEE Transactions on Magnetics 2003, 39, 295-304.

139. Karden, E.; Ploumen, S.; Fricke, B.; Miller, T.; Snyder, K. Energy storage devices for future hybrid electric vehicles. Journal of Power Sources 2007, 168, 2-11.

140. Zhang, G.; Li, H.; Antensteiner, M.; Chung, T. C. M. Synthesis of Functional Polypropylene Containing Hindered Phenol Stabilizers and Applications in Metallized Polymer Film Capacitors. Macromolecules 2015, 48, 2925-2934.

141. Ma, R.; Sharma, V.; Baldwin, A. F.; Tefferi, M.; Offenbach, I.; Cakmak, M.; Weiss, R.; Cao, Y.; Ramprasad, R.; Sotzing, G. A. Rational design and synthesis of polythioureas as capacitor dielectrics. Journal of Materials Chemistry A 2015, 3, 14845-14852.

142. Offenbach, I.; Gupta, S.; Chung, T. C. M.; Weiss, R. A.; Cakmak, M. Real-Time Infrared−Mechano-Optical Behavior and Structural Evolution of Polypropylene and

200

Hydroxyl-Functionalized Polypropylene during Uniaxial Deformation. Macromolecules 2015, 48, 6294-6305.

143. M, C.; M., S. Dynamics of Uni and Biaxial Deformation and Its Effects on the Thickness Uniformity and Surface Roughness of Poly (Ether Ether Ketone) Films. You have full text access to this content 2004, 35 (19), 1562-1568.

144. Tubes, R. K. Melting Point and Heat of Fusion of Poly(viny1 Alcohol). Journal of Polymer Science Part A: 1956, 3, 4181-4189.

145. Roberts, C. Poly(ethylene terephthalate) I Heat of Fusion. Polymer 1969, 10, 113- 116.

146. Jordan, E. F.; Veldeisen, D. W.; Wrigley, A. N. Side-Chain Crystallinity. I. Heats of Fusion and Melting Transitions on Selected Homopolymers Having Long Side Chains. Journal of Polymer Science Part A 1971, 9, 1835-1852.

147. Flory, J. P. Principles of Polymer Chemistry. Cornell University Press: Ithaca, 1953; pp 568-570.

148. Martins, C. I.; Cakmak, M. Role of Relaxation on Strain Crystallization of Uniaxially Stretched Poly(ethylene naphthalate) Films: A Mechanooptical Study. Macromolecules 2006, 39, 4824– 4833.

149. Yang, Y.; Yin, D.; Xiong, R.; Shi, J.; Tian, F.; Wang, X.; Lei, Q. FTIR and Dielectric Studies of Electrical Aging in Polyimide under AC Voltage. IEEE Transactions on Dielectrics and Electrical Insulation 2012, 19 (2), 574-580.

150. Ha, K.; West, J. L. Studies on the Photodegradation of Polarized UV-Exposed PMDA–ODA Polyimide Films. Journal of Applied Polymer Science 2002, 86, 3072- 3077.

201

151. Cakmak, M.; Hassan, M.; Unsal, E.; Martins, C. A fast real time measurement system to track in and out of plane optical retardation/ birefringence, true stress, and true strain during biaxial stretching of polymer films. Review of Scientific Instruments 2012, 83, 123901.

152. M, C.; L., W. J.; E., S. J. Optical properties of simultaneous biaxially stretched poly (ethylene terephthalate) films. Polymer Engineering & Science 1989, 19 (21), 1534- 1543.

202