1 SYNTHESIS AND CRYSTALLIZATION OF PRECISE AND RANDOM HALOGEN CONTAINING POLYOLEFINS By EMINE BOZ A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR TH E DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2007
2 2007 Emine Boz
3 For Dr. Butler
4 ACKNOWLEDGMENTS It is difficult to express how much my four years in graduate school have impacted my life. During this time there have been so many different experiences and challenges, which I could not have overcome without the help, guidance, and frien dship of the very important people in my life. Most of all I thank my advisor, Prof. Kenneth B. Wagener. It is because of Dr. Wagener that I came to UF to pursue graduate studies and it is because of his constant support and positive influence that I have been able to continue and complete my education here. Dr. Wagener has shown me what it means to be a person of character and a person who truly cares from his heart about the well being of others. He is much more than an accomplished scientist and advisor, he is a source of lasting influence and inspiration. I also thank the number of excellent scientists with whom I have had an opportunity to work. Prof. Rufina G. Alamo from Florida State University played a large role in the development of my project and taught me much to increase my knowledge and understanding. Working with Dr. Alamo has been a very valuable experience. Dr. Ion Ghivriga has also been extremely helpful in offering his expertise and time in the characterization of many of my polymer sample s. Dr. Lisa Baugh, at Exxon Mobil, has also been of great help in assisting with polymer characterization and has always been so enthusiastic and friendly. Coworkers in the lab have also played a large role in my time here at UF. Most importantly is Alexan der J. Nemeth, perhaps the most exceptional undergraduate I have ever met. Alex worked alongside me in the lab and was always eager to put in long hours in order to make sure that the experiments were done right. I will miss our teamwork and all the conver sations about baseball. I also thank Dr. John C. Sworen for his guidance during the early days of my research. John was an awesome person to work with. Dr. Tim Hopkins was also a very helpful influence in
5 the lab. I would also like to thank the labmates an d group members who made my time at UF an enjoyable and productive experience. Thanks go also to the members of my committee. Prof. William Dolbier was not only a member of my committee; he was also my teacher, and a helpful source of great knowledge. I a spire to one day be the teacher that he is. Prof. Anthony Brennan is the coolest professor I have ever met. I enjoyed all the chances I had to interact with him. Prof. Adam Viege was also a very helpful member of my committee. Special thanks go to Prof. Ro nald Castellano and Prof. Lisa McElwee White for serving on my committee at short notice on two different occasions. Finally, I thank Prof. Merle Battiste, it was great experience to TA for him and a pleasure to discuss science with him. I must also thank the angels of the polymer floor: Lorraine Williams, Sara Klossner, and Gena Borrero. Without their patient and kind help on a daily basis life would be so much more difficult. I will miss our conversations in the polymer office. I also thank my two closest friends on the polymer floor, Christophe Grenier and Barry Thompson. I do not have the words to express my thanks and appreciation for my family. They have been totally supportive of me throughout my life and education. They have thought only of what wil l be best for me and make me happy. Their support cannot be quantified and even though they are far away, I can feel their love and support at every moment. Finally, I need to thank the person who made my education at UF possible and that is Prof. George Butler. I am grateful that I had the opportunity to meet Dr. Butler and to get to know him, even if for just a short time. It is people like Dr. Butler that make the hopes and ambitions of others possible. It is hard to imagine the number of people that he has influenced in a positive
6 way through the legacy that he has established at UF. I hope that I can one day continue his legacy, which has been passed down to me through Dr. Wagener.
7 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ............... 4 LIST OF TABLES ................................ ................................ ................................ ........................... 9 LIST OF FIGURES ................................ ................................ ................................ ....................... 10 ABSTRACT ................................ ................................ ................................ ................................ ... 13 CHAPTER 1 PROGRESS IN THE DEVELOPMENT OF WELL DEFINED ETHYLENE VINYL HALIDE POLYMERS ................................ ................................ ................................ ........... 15 1.1 Introduction ................................ ................................ ................................ ....................... 15 1.2 Halogen Containing Homopolymers ................................ ................................ ................ 16 1.2.1 Poly(vinyl halides) ................................ ................................ ................................ .. 16 188.8.131.52 Poly(vinyl fluoride) ................................ ................................ ...................... 16 184.108.40.206 Poly(vinyl chloride) ................................ ................................ ...................... 18 220.127.116.11 Poly(vinyl b romide) ................................ ................................ ..................... 20 18.104.22.168 Poly(vinyl iodide) ................................ ................................ ......................... 21 1.2.2 Poly(vinylidene halides) ................................ ................................ ......................... 22 22.214.171.124 Poly(vinylidene fluoride) ................................ ................................ ............. 22 126.96.36.199 Poly(vinylidene chloride) ................................ ................................ ............. 24 188.8.131.52 Poly(vinylidene bromide) ................................ ................................ ............. 25 1.2.3 Polytetrafluoroethylene ................................ ................................ .......................... 26 1.3 Ethylene Vinyl Halide (EVH) Copolymers ................................ ................................ ...... 28 1.3.1 Halogenation of Polyethylene ................................ ................................ ................ 29 184.108.40.206 Fluorinated PE ................................ ................................ .............................. 29 220.127.116.11 Chlorinated PE ................................ ................................ ............................. 30 18.104.22.168 Brominated PE ................................ ................................ ............................. 35 1.3.2 Dehalogenated Homopolymers ................................ ................................ .............. 37 1.3.3 Synthesis of EVH Copolymers via Ring Op ening Metathesis Polymerization ..... 39 1.4 Acyclic Diene Metathesis (ADMET) Polymerization ................................ ...................... 40 2 SYNTHESIS AND CRYSTALLIZ ATION OF PRECISION ADMET POLYOLEFINS CONTAINING HALOGENS ................................ ................................ ................................ 45 2.1 Introduction ................................ ................................ ................................ ....................... 45 2.2 Results and Discussion ................................ ................................ ................................ ..... 49 2.2.1 Synthesis ................................ ................................ ................................ ................. 49 2.2.2 Primary Structure Characterization. ................................ ................................ ....... 50 2.2.3 Wid e Angle X Ray Diffraction ................................ ................................ .............. 51 2.2.4 Melting and Crystallization Behavior ................................ ................................ .... 55 2.2.5 Solid State Nuclear Magnetic Resonance ................................ .............................. 56
8 2.3 Conclusions ................................ ................................ ................................ ....................... 62 2.4 Experimental Section ................................ ................................ ................................ ........ 63 3 PRECISION ETHYLENE / VINYL CH LORIDE POLYMERS VIA CONDENSATION POLYMERIZATION ................................ ................................ ............. 79 3.1 Introduction ................................ ................................ ................................ ....................... 79 3.2 Results and Discussion ................................ ................................ ................................ ..... 82 3.2.1 Monomer and Polymer Synthesis ................................ ................................ ........... 82 3.2.2 Primary Structure Characterization ................................ ................................ ........ 83 3.2.3 Th ermal Analysis ................................ ................................ ................................ .... 87 3.2.4 Differential Scanning Calorimetry ................................ ................................ ......... 89 3.3 Conclusion and Outlook ................................ ................................ ................................ ... 90 3.4 Experimental ................................ ................................ ................................ ..................... 91 4 PRECISION ETHYLENE/VINYL BROMIDE POLYMERS ................................ ............ 103 4.1 Introduction ................................ ................................ ................................ ..................... 103 4.2 Results and Discussion ................................ ................................ ................................ ... 104 4.3 Conclusion ................................ ................................ ................................ ...................... 107 4.4 Experimental ................................ ................................ ................................ ................... 108 5 WELL DEFINED PRECISION ETHYLENE/VINYL FLUORIDE POLYMERS VIA CONDENSATION POLYMERIZATION ................................ ................................ ........... 117 5.1 Introduction ................................ ................................ ................................ ..................... 117 5.2 Results and Discussion ................................ ................................ ................................ ... 118 5.2.1 Monomer and Polymer Synthesis ................................ ................................ ......... 118 5.2.2 Prima ry Structure Characterization ................................ ................................ ...... 118 5.2.3 Thermal and X Ray Analysis ................................ ................................ ............... 1 19 5.3 Conclusions ................................ ................................ ................................ ..................... 122 5.4 Experimental ................................ ................................ ................................ ................... 122 6 STATISTICALLY RANDOM, DEFECT FREE ETHYLENE / VINYL HALIDE MODEL COPOLYMERS VIA CONDENSATION POLYMERIZATION ....................... 132 6.1 Introduction ................................ ................................ ................................ ..................... 132 6.2 Results and Discussion ................................ ................................ ................................ ... 133 6.2.1 Polymer Synthesis ................................ ................................ ................................ 133 6.2.2 Primary Structure Characterization ................................ ................................ ...... 134 6.2.3 Thermal Analysis ................................ ................................ ................................ .. 138 6.2.4 D ifferential Scanning Calorimetry ................................ ................................ ....... 139 6.3 Conclusion and Outlook ................................ ................................ ................................ 142 6.4 Experimental ................................ ................................ ................................ ................... 142 LIST OF REFERENCES ................................ ................................ ................................ ............. 159 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ....... 170
9 LIST OF TABLES Table page 1 1 Properties of poly(vinyl halide) polymers. ................................ ................................ ........ 42 1 2 Properties of poly(vinylidene halide) p olymers. ................................ ................................ 42 2 1 Crystalline data of precision ADMET polyolefins (PE19X) ................................ ............ 71 2 2 Lattice parameters from WAXD for orthorhombic PE 19X (X = H, O, F) crystals. ......... 71 2 3 T 1 H estimations from two component fits for PE19X (X = H, F, Cl, Br) ........................ 71 2 4 13 C NMR chemic al shifts for crystalline carbons of PE19X precision ADMET polyolefi ns ................................ ................................ ................................ .......................... 72 3 1 Molecular weight and thermal data for unsaturated (UPEXCl) and saturated (PEXCl) polymers. ................................ ................................ ................................ ............................ 96 3 2 Proton and carbon chemical shifts (in ppm) and trans cis (in ppb) for polymers UPE9Cl UPE15Cl and UPE21Cl ................................ ................................ .................... 96 4 1 Summary of thermal propert ies measured via DSC for the family of ethylene/vinyl bromide polymers. ................................ ................................ ................................ ........... 113 5 1 Molar mass and thermal data of precision EVF ................................ ............................... 127 5 2 Orthorhombic lattice parameters and crystallinity from WAXD patterns ....................... 127 6 1 Polymer p roperties of precise and random ADMET samples ................................ ......... 148 6 2 Monomer ratio, degree of randomness (DR) and trans : cis ratio in copolymers. ............ 148 6 3 13 C chemical shifts in random copolymers. ................................ ................................ ..... 148 6 4 Differences in 13 C chemical shifts between the Cl, Br, and F copolymers. ..................... 149 6 5 Differences in 13 C chemical shifts between the trans and cis configuration of the double bond in copolymers. ................................ ................................ ............................. 149 6 6 Differences in 13 C chemical shifts t and c in copolymers. ................................ ....... 149 6 7 Increments for the calculus of the 13 C chemical shifts in copolymers. ............................ 150
10 LIST OF FIGURES Figure page 1 1 Structure of poly(vinyl halide) polymers. ................................ ................................ .......... 43 1 2 Unit cell parameters ................................ ................................ ................................ .......... 43 1 3 Structures of poly(vinylidene halide) polymers. ................................ ................................ 44 1 4 Routes to EVC polymers. ................................ ................................ ................................ .. 44 2 1 Monomer and polymer s ynthesis. ................................ ................................ ...................... 72 2 2 TGA for PE19F (a) and PE19Br ................................ ................................ ........................ 73 2 3 IR spectra for thin films of PE19F, PE19 Cl, and PE19Br cast on KBr disks. ................... 74 2 4 WAXD diffractograms of linear PE and ADMET precisely substi tuted polyethylenes ................................ ................................ ................................ .................... 75 2 5 DSC exotherms (a) and endotherms (b) of linear PE and precis ely substituted polyethylenes ................................ ................................ ................................ ..................... 76 2 6 Peak melting temperatures of precisely substituted polyethylenes (PE19X) versus van der W aals r adius of substituent (X) ................................ ................................ ............. 76 2 7 Fully relaxed 19 F (a) and 13 C NMR (b, c) DD/MAS spectra of PE19F PE19Cl and PE19Br respectively ................................ ................................ ................................ ........... 77 2 8 Solid state 13 C NMR. (a) Unfiltered 13 C NMR CP MAS spectra and (b) 1 H spin locking filtered (crystalline) spectra of precision PE19X (X = F, Cl, Br) polyolefins. ..... 78 3 1 Synthesis of PE XCl polymers. ................................ ................................ ........................... 97 3 2 1 H NMR spectra for monomer 7 (a), UPE9Cl (b), PE9Cl (c) and 13 C NMR spectra for monomer 7 (a), UPE9Cl (b), PE9Cl (c). ................................ ................................ ...... 98 3 3 IR of PE, PE21Cl, PE15Cl, PE9Cl, and PVC. ................................ ................................ ... 99 3 4 TGA results for PE, PE2 1Cl, PE15Cl, PE9Cl, and PVC ................................ ................. 100 3 5 DC S exotherms (a) and endotherms (b) of PE9Cl, PE15Cl, PE21Cl, relative to PE and PVC. ................................ ................................ ................................ .......................... 101 3 6 DSC endotherms for PE9Cl at heating rates of 5, 10, and 40 C/min. ............................ 101 3 7 Trends for the variation in T m vs. the number of CH 2 per repeat unit. ............................ 102
11 4 1 Synthesis of precision bromine polymers. ................................ ................................ ....... 113 4 2 Thermogravimetric analysis and infrared spec tra of PEXBr ................................ ........... 114 4 3 Melting and heat of fusion tre nds for PEXBr polymers ................................ .................. 115 4 4 IR spectra of U PE9Br, UPE15Br, UPE21Br ................................ ................................ ... 116 4 5 T m vs number of moles of CH 3 in repeating unit for PE9CH 3 PE11CH 3 PE15CH 3 PE19CH3, PE21CH 3 and PE. ................................ ................................ ......................... 116 5 1 Synthesis of precision EVF polymers. ................................ ................................ ............. 127 5 2 Infrared spectra of PE, PE21F, PE15F, PE9F. ................................ ................................ 128 5 3 Thermogravimetric analysis results for PE21F, PE15F, and PE9F. ................................ 128 5 4 DSC exotherms and endotherms for PE9Cl, PE15Cl, PE21Cl. ................................ ....... 129 5 5 W AXD of a linear polyethylene fraction and ADMET precision EVF samples ............. 130 5 6 Trends for the variation in T m (a) and H f (b) vs. mole % halogen per repeating unit. .. 131 6 1 Synthesis of random copolymers via ADMET. ................................ ............................... 150 6 2 Repeat unit structure f or RPEXCl polymers. ................................ ................................ ... 150 6 3 1 H NMR spectra for RUPE15Cl (top) and RUPE21Cl (bottom). ................................ .... 150 6 4 13 C NMR for RUPE15Cl (top) and RUP E21Cl (bottom). ................................ .............. 151 6 5 Expansion (1) of the GHMBC spectrum of RUPE15Cl. ................................ ................. 151 6 6 Expansion (2) of the GHMBC spectrum of RUPE15Cl ................................ ................. 152 6 7 Expansion (3) of the GHMBC spectrum of RUPE15Cl. ................................ ................. 152 6 8 13 C NMR for RUPE15Cl (top) and RUPE21Cl (bottom). ................................ .............. 153 6 9 Expansion (4) of the GHMBC spectrum of RUPE15Cl. ................................ ................. 153 6 10 13 C NMR for RUPE15Cl (top) and RUPE21Cl (bottom). ................................ .............. 154 6 11 Expansion (5) of the GHMBC spectrum of RUPE15Cl. ................................ ................. 154 6 12 Expansion (6) of the GHMBC spectrum of RUPE15Cl. ................................ ................. 155 6 13 IR spectra for RPE15F (top), RPE15Cl (middle), and RPE15Br (bottom). .................... 155 6 14 TGA results for all six random polymers ................................ ................................ ....... 156
12 6 15 DSC comparison of random copolymers and precise analogu es ................................ ..... 157 6 16 Melting thermograms of precise PE21Cl (a ) and random copolymer analogue RPE21Cl (b), isothermally crystallized at 7 6 C and 7 3 C respectively for the times indicated. ................................ ................................ ................................ .......................... 158
13 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SYNTHESIS AND CRYSTALLIZATION OF PRECISE AND RANDOM HALOGEN CONTAINING POLYOLEFINS By Emine Boz August 2007 Chair: Kenneth B. Wagener Major: Chemistry We report the synthesis and characterization of a series of precision polyethylene (PE) structures containing a fluorine, chlorine, or bromine on each and every 9 th 15 th 1 9 th and 21 st carbon. The use of acyclic diene metathesis (ADMET) polymerization allows unprecedented control in the synthesis of these halogen containing PE derivatives, and results in the first bromine containing polyolefins with precisely defined primar y structures. ADMET polymerization followed by hydrogenation leads to these precise structures, which have been characterized by 1 H NMR, 13 C NMR, IR, elemental analysis, TGA, DSC, and WAXD. The TGA data, coupled with elemental analysis, supply definitive p roof of the structural composition through the observed thermal decomposition and release of exact masses of HX (X = F, Cl, or Br). Relative to analogous random copolymers, these precisely substituted polymers display sharper WAXD diffraction patterns, hig her crystallinities and much narrower melting peaks, typical of homopolymer crystallization. This crystallization behavior is supported by solid state NMR based on the observed equivalence in the relative distribution of halogens between crystalline and no n crystalline regions. Lattice distortions caused by the accommodation of the substituent in the lattice, render a change from orthorhombic to triclinic structures at a van der Waals atomic radius of the substituent > 1.6 The observed melting and enthal pies of fusion
14 decrease dramatically with increasing volume of the substituent and scale proportionally to the van der Waals atomic radius in the halogenated series. The synthesis of ethylene vinyl halide (EVH) copolymers containing fluorine, chlorine, a nd bromine via the ADMET copolymerization of halogen containing dienes with 1,9 decadiene is also presented. Analysis of the unsaturated prepolymers via 13 C NMR spectroscopy establishes the statistically random nature of the copolymers. Thermal analysi s indicates a distinct difference in the crystallization behavior of these random copolymers when compared to their compositionally matched precise analogues. Here, the homopolymer crystallization is no longer observed and is replaced by a mechanism based on the selection of long c rystallizable sequences. The results presented thus point to the utility of these random ADMET copolymers as suitable models for industrially relevant PE copolymers based on the perfectly linear, defect free, and statistically random copolymer composition.
15 CHAPTER 1 P ROGRESS IN THE DEVEL OPMENT OF WELL DEFINED ETHYLENE VINYL HALIDE POLYMERS 1.1 Introduction Polyethylene (PE) is the largest volume polymer produced in the world today. 1 The range of applications for PE is so diverse, such that it is difficult to adequately describe. The real strength of PE is that a vast property set is available, as determined by the structural characteristics of a specific sample, which are tuned in the employed synthetic process. While variation in synthesis and processing con ditions is used to target PE for given applications, the available range of properties is vastly increased through the use of polymer additives 2 or via the introduction of chemical functionality to the parent PE backbone. 3 The chemical industry has long be en dominated by the additive approach, based on the simplicity and tunability of this technique. However, polymer additives do present drawbacks and fail to modify the polymer structure at its most fundamental level. 4 Advances in polymerization techniques have now made it possible to gain an ever greater control over polymer structure at all levels, enabling the exploration of property control strictly via manipulation of macromolecular structure. Such an approach not only allows the precise tuning of prop erties in a single, well defined material, but it also allows one to learn the fundamental consequences of structural modification. The strong effects of structural modification are clearly seen in the great variety of halogen containing PE analogues, in w hich the halogen type, content, and sequence distribution, collectively modulate the structural and physical properties of the polymer. In this chapter, a critical survey of halogen containing polymers is presented, with the goal of illustrating what is k nown about the role of halogen substituents on the properties of linear PE based analogues. In order to focus on developing direct correlations between primary structure
16 and bulk properties, only those halogen containing polymers with just one type of halo gen per sample will be considered. In the first section, industrially relevant homopolymers are described in order to establish a base point for the types of properties exhibited by these materials and the role of the halogen substituent in defining these properties. In the second section, the most common classes of chemically modified halogen containing PE analogues are examined for the structure property relationships that can be derived from judicious structural modification. 1.2 Halogen Containing Hom opolymers 1.2.1 Poly(vinyl halides) The simplest halogen containing iteration of PE is the family of poly(vinyl halide) (PVH) polymers shown in Figure 1 1: poly(vinyl fluoride) (PVF), poly(vinyl chloride) (PVC), poly(vinyl bromide) (PVB), and poly(vinyl i odide) (PVI). By comparing the fundamental properties of the polymers in this family, the effect of halogen type on the structural properties of these PE analogues can be understood. In all cases (as with the PE), the effect of the synthetic method cannot be ignored, as it exerts a strong influence on the polymer structure. Here, the properties presented for a given polymer will be those that result from the most common synthetic route for that specific polymer. 22.214.171.124 Poly(vinyl fluoride) Poly(vinyl fl uoride) (PVF) is represented by the repeat unit shown in Figure 1 1. Here the PE backbone has been modified only by the addition of a fluorine atom on every other carbon. While the size of fluorine is only slightly larger than that of hydrogen (van der Waa ls radius = 1.47 vs. 1.2 for H ) the strong electronegativity difference between the C H bond dipole and the C F bond dipole is not negligible and plays a role in the overall structure of the polymer. The other factor that is introduced by addition of the halogen in PVF (and all of the PVH polymers) is tacticity and whether the polymer is atactic or possesses any degree of stereoregularity.
17 The synthesis of PVF is achieved most commonly by radical initiated suspension polymerization of vinyl fluoride, 5 although emulsion polymerization can also be used. 6 Vinyl fluoride is the most difficult vinyl halide monomer to polymerize. The result is that very high pressures are required. 7 Typically PVF contains about 88 90% head to tail linkages as is represented by the repeat unit, although the presence of head to head linkages is a consequence of the free radical polymerization method. 8 The polymer is atactic, based on the polymerization condition used. For PVF, the molecular weight is generally reported as a vis cosity average, which is approximately 60,000 in DMF at 90 C. 9 PVF is a semicrystalline polymer, which has a percent crystallinity (X c ) of 20 60% depending on the synthetic method used. 10 The melting temperature (T m ) for PVF is typically observed at 190 C for commercial samples, which is ~60 C higher than that observed for PE, 11 12 13 reflecting the stronger nature of the interactions within the crystallites in PVF. Two distinct glass transitions (T g ) are observed, which emphasize the semicrystalline nature of the polymer. The lower T g occurs at 15 to 20 C and corresponds to the relaxation of amorphous domains, which are free from constraint by crystallites and the upper T g occurs at 40 50 C and corresponds to the relaxation of amorphous domains constrain ed by crystallites. 14 Overall the polymer has a greater tendency to crystallize than other PVH polymers, which is due partly to the smaller size of the fluorine atom as well as to the high electronegativity of the fluorine, which induces strong dipole dipol e interactions (hydrogen bonding), enforcing a more crystalline structure. However, the significant variation in crystallinity (20 60%) is thought to be due primarily to defect structures, such as head to head linkages and branching but not to variations i s tacticity. 10
18 An orthorhombic crystal structure has been proposed for PVF, with unit cell dimensions: a = 8.57 b = 4.95 and c = 2.52 15 16 For comparison, PE is also found to crystallize with an orthorho mbic structure and dimensions: a = 7 .42 b = 4 .94 and c = 2 55 17 18 19 The discrepancy in the unit cell dimensions, found primarily in the expansion of the lattice in the a dimension, reflects the larger size of the fluorine atom relative to hydrogen. Figure 1 2 illustrates the definition of the unit cell for several common unit cell types discussed in this work and pictorially shows the unit cell for PE as a representative structural example in order to visualize how polymers organize in a crystalline lattice. PVF is more thermally stable than other vinyl halide polymers and high molecular weight PVF is reported to degrade in an inert atmosphere, with concurrent HF loss and backbone cleavage occurring at about 450 C. 20 In air, HF loss occurs at about 350C, followed by backbone cleavage around 460C. The excellent thermal stability is due to the large number of hydrogen bonds. This high stability is also reflected in several other physical properties of PVF, such as excellent resistance to sunlight deg radation, chemical attack, water absorption, and solvent, which impart excellent outdoor durability and stain resistance. 7 This property set, coupled with a high degree of transparency to visible light, leads t o the application of PVF primarily as a protective coating in a variety of outdoor and indoor settings, specifically for: buildings, trucks, trains, airplanes, greenhouses, solar cells, fuel lines, and circuit boards. 7 126.96.36.199 Poly(vinyl chloride) Poly(vinyl chloride) (PVC) (Figure 1 1) is the second largest volume plastic produced in industry and its relatively low cost and attractive property set lead it to be one of the most versatile and applicable polymer s known. 21 PVC is synthesized via free radical suspension, emulsion, or bulk polymerization, with suspension methods accounting for 80% of the polymer produced worldwide. 22 Typical PVC has a degree of polymerization of 500 3500, corresponding
19 to molecular we ights in the range of ~30 200 kDa. 21 Molecular weight can be tuned via reaction temperature, with higher temperatures producing lower molecular weight polymers. PVC shows a slight tendency toward a syndiotactic structure, although polymerization temperature affects the syndiotactic content, with lower temperatures leading to higher syndiotactic content. A typical sample polymerized at 50 C (a typical industrial polymerization temperature) is 56% syndiotactic. 23 The most important structural consequence of the free radical polymerization process used is the high defect content in PVC. While the primary structure of PVC is e ssentially defined by head to tail linkages, the polymer contains many structural defects c onsisting of lon g and short chains terminal and internal unsaturation branching, and end group defects 24 In bulk samples, PVC is a semicrystalline polymer that is largely amorphous. Samples are generally less than 10% crystalline, which is primarily att ributed to irregular structure, defined by random distributions of syndiotactic sequences of varied length. 25 The effect of this irregular structure is to inhibit the crystallization of the polymer based on the difficulties of like segments to find each oth er and pack into crystallites. The crystallites in PVC are lamellar in nature and have been assigned an orthorhombic structure with unit cell dimensions a = 10.6 b = 5.4 c = 5.1 26 However, the wide range of crystallite sizes and the generally imper fect nature of the crystallites leads to a broad melting temperature for PVC of ~120 260 C. 27 The T g of PVC is 82 C, which leads it to be a brittle material at room temperature. 28 The most important thermal property of PVC is related to the relative instab ility of PVC at elevated temperatures. Even at 100 C PVC is known to undergo a very slow elimination of the labile chlorine atoms in the form of HCl. As the temperature is increased, the rate of HCl evolution is greatly enhanced at about 200 C. This elim ination process leads to the formation of
20 unsaturated sites in the polymer backbone and is characterized as the first of three steps in the thermal degredation process of PVC. 29 As double bonds are generated in the backbone, neighboring chlorine atoms becom e allylic chlorines, and are subsequently even more labile to further elimination. Such a process rapidly turns into a chain reaction characterized as the unzipping mechanism. Defect sites serve as initiation centers for this mechanism, as tertiary and all ylic chlorines are much more easily eliminated at lower temperatures. The consequence of this first mechanistic stage of decomposition is a polyene structure, which is unstable to a variety of further reactions defining stage two of decomposition, which oc curs at temperatures greater than ~350 C. In this case, the relatively electron rich polyene structure can undergo oxidative chain scission or crosslinking reactions, such as Diels Alder cycloadditions. The final stage of decomposition is that of char for mation, in which aromatic hydrocarbons generated by extensive cross linking and chain scission are released. However, the cross linked mass formed in step two is never fully decomposed and remains even at extremely high temperatures, above 500 C. The phy sical properties of PVC are generally widely tuned via the addition of additives and plasticizers, which can generate a range of mechanical properties in the polymer, from a rigid form to a highly flexible form. The overall good mechanical properties, coup led with a strong chemical resistance (except in the presence of low molecular weight chlorinated solvents) make PVC applicable in a diverse range of settings, which include: pipes, siding, flooring, roofing, wall covering, chemical plant construction mate rials, flexible wire and cable covering. 30 188.8.131.52 Poly(vinyl bromide) Poly(vinyl bromide) (PVB) (Figure 1 1) is significantly less well known than its analogues PVF and PVC. The main reason for the lack of application stems from the lower stability of PV B, induced by the significantly more labile C Br bond, which renders the polymer unstable to light and heat. 31 PVB is typically synthesized by free radical methods in bulk,
21 solution, suspension, or emulsion, using typical peroxides or AIBN for initiation. 31 The polymer is atactic when synthesized at room temperature, but at lower polymerization temperatures tends to show increased syndiotactic character. 32 No general property set is available for a so called range of 20 50 kDa have been reported for PVB. 33 The most universal property of PVB is the relative thermal instability of the polymer. Monitoring by TGA, PVB has been reported to show two distinct thermal degradation processes. The first is the loss of HBr, which reaches a maximum rate at ~100 C (although temperatures from 100 200 C have been reported). 34 The second stage decomposition reaches a maximum rate at ~500 C and corresponds to the crosslinking and fragmentation of the polyene structure that is formed during the evolution of HBr. Due to the low decomposition temperature of PVB, relatively little additional thermal data is reported. The T g of PVB has been repo rted to range from 50 100 C. 28 Only a few studies have focused on the crystalline properties of PVB as measured by XRD, 35 assignment to an orthorhombic unit cell is reported 36 with dimensions a = 11.0 b = 5.6 c = 5.1 but no percent crystallinity is given. 184.108.40.206 Poly(vinyl iodide) (PVI) Poly(vinyl iodide) (PVI) (Figure 1 1) is even more unstable than PVB and consequently, even less literature is found concerning this polymer, which has no known applic ations. Reported syntheses of PVI include the gamma ray induced radical polymerization of vinyl iodide and the AIBN initiated radical polymerization. 37 38 Sodium thiosulphate (Na 2 S 2 O 3 ) is an important additive to polymerizations of vinyl iodide, which acts t o consume inhibitors such as iodine and hydrogen iodide, which are produced during the polymerization as the decomposition products of the vinyl iodide monomer. 39 Polymer molecular weights on the order of 4500 7500 have been
22 reported, but the polymer is the rmally unstable and decomposes at less than 100 C and is also unstable to light. 40 Table 1 1 summarizes the properties of the PVH polymers discussed. It can be observed that although the orthorhombic structure of PE is maintained throughout the series, th e dimensions of the unit cell increase with increasing halogen size. A concurrent decrease in T m and percent crystallinity with increasing halogen size is also observed. 1.2.2 Poly(vinylidene halides) Vinylidene halide polymers are the next logical itera tion of halogen containing polyolefins and are represented by the repeat unit (CH 2 CX 2 ) and the common structures are shown in Figure 1 3 With two halogens on every other carbon, tacticity is not a concern in the vinylidene halide family. This homopolym er family provides a deeper insight into the effects of halogen content, when compared to the vinyl halide polymers, as well as the effects of halogen type when compared across the two full series. 220.127.116.11 Poly(vinylidene f luoride) (PVDF) The synthesis o f PVDF is achieved by radical polymerization in either emulsion or suspension of the gaseous monomer 1,1 difluoroethylene. The polymerization is generally performed in water at temperatures up to 150 C and pressures up to 100 MPa with organic peroxides as the initiators. The degree of polymerization of industrial PVDF is usually around 1500, giving a molecular weight of 50,000 kDa (M n PDI =2) on average. 41 By NMR it can be observed that between 2% and 7% of head to head and tail to tail links 42 are found in typical PVDF and that the defect level increases with increasing polymerization temperature. 10 T he defect linkages affect significantly the crystallization processes and the resulting morphology. 43 44
23 PVDF is kn own to have three major crystalline forms, which find their basis in three different chain conformations, which are abbreviated as and The form is thermodynamically the most favorable because the F F interactions are minimized by the trans gauch e conformation. 45 When PVDF is crystallized from the melt at room temperature, the form dominates and results in polymer samples that are 50 70% crystalline. 46 The fundamental crystalline nature of such polymers is found in an orthorhombic unit cell with d imensions: a = 4.96 b = 9.64 c = 4.62 47 The T m of the phase is measured to be 172 C. One of the most important consequences of the phase is that the strong dipole moments of the CF 2 units are aligned antiparallel in adjacent chains. If the cry stallization from the melt is carried out at higher temperatures (>160 C), a new phase is observed. In this case a monoclinic unit cell is observed with dimensions: a = 4.96 b = 9.67 c = 9.20 48 The T m of the phase is 185 C. A modified chain c onformation relative to the phase is observed and characterized by a trans trans trans gauche conformation in which the dipole moments are now aligned in adjacent chains and do not cancel. 49 The third common form is the phase, which is characterized by an all trans polymer backbone, which crystallizes with an orthorhombic unit cell with dimensions: a = 8.58 b = 4.91 c = 2.56 The T m of the phase is 178 C. 49 In the phase all the dipole moments are aligned parallel and thus this phase is the most interesting for electronic applications, as will be described below, based on the unusually high dielectric constant of this material. 50 Formation of the phase is usually negligible in the melt crystallize d samples, as it is not nearly as kinetically favorable as the and phases. The phase is usually derived by drawing of phase samples at temperatures > 100 C. The T g of the amorphous regions of PVDF is observed at 40 30 C. 51 52 53
24 The thermal stab ility of PVDF is excellent up to 300 C, at which point the onset of thermal degradation can be observed. At temperatures above 350 C, a strong evolution of HF is observed. 46 The thermal stability of PVDF, cou pled with the stiffness and toughness induced by the high crystalline content of the polymer, make PVDF important in applications in which toughness, impact resistance, and mechanical strength are required. PVDF is also resistant to chemicals, weathering, and light. The high dielectric constant, especially of the phase, makes PVDF important for pyroelectric and piezoelectric applications. 41 Common uses of PVDF include coatings, laminates, tubing, piping, jacke ting for wires, and piezo and pyroelectric transducers. 18.104.22.168 Poly(vinylidene chlori de) (PVDC ) While vinylidene chloride copolymers have long been well known as among the first commercially available polymers, the homopolymer PVDC does not have suffic ient thermal stability and is therefore not commercially available. 54 Nonetheless, from a structural point of view, PVDC is interesting for comparison purposes. PVDC is synthesized by a variety of free radical methods in solution, suspension, emulsion, or s lurry, initiated by peroxides or azo compounds, resulting in a primarily head to tail polymer, which contains relatively little head to head defects or branching. 55 Degrees of polymerization from 100 10,000 are readily achieved in PVDC. 56 Due to the highly r egular structure that is found with PVDC, a high degree of crystallinity is observed, similar to that described for PVDF above. The physical properties of PVDC as observed by DSC, reflect the highly crystalline nature of the polymer, which shows only a v ery weak T g at 19 11 C in the as polymerized samples. 57 58 Polymer melting is observed at 198 205 C. 59 60 The unit cell as measured by XRD
25 of PVDC is defined by dimensions: a = 6.73 b = 12.54 c = 4.68 and is assigned a monoclinic unit cell. 61 The degradation behavior of PVDC is its most well known property. At temperatures above 120 C, PVDC begins to undergo dehydrohalogenation and a rapid evolution of HCl gas is observed just above the T m at ~200 C. 62 This lack of thermal stability renders PV DC commercially unviable. 22.214.171.124 Poly(vinylidene b romide) (PVDB) Polymerization of PVDB is achieved by emulsion polymerization of vinylidene bromide. 63 As with PVB, PVDB is unstable to light and is thermally unstable. The onset of degradation for PVDB occ urs at a slightly lower temperature than for PVB, but the release rate of HBr is not as rapid as for PVB. 64 Because of the relative instability of PVDB, little information exists concerning this polymer. One report does compare the observed crystal structur e of PVDC and PVDB and gives unit cell dimensions for a monoclinic unit cell of PVDB of: a = 25.88 b = 13.87 c = 4.77 and directly compared to PVDC, which was found to have unit cell dimensions: a = 22.54 b = 12.53 c = 4.68 65 The b and c d imensions reported here for PVDC match closely with those reported in the previous section. Notice here, that the a dimension for both PVDC and PVDB are considerably larger than the 6.73 reported in the previous section for PVDC and reflect a different d efinition of the unit cell. 61 Importantly though, the noticeable expansion in the unit cell size when going from PVDC to PVDB is a reflection of the incorporation of the larger bromine atoms into the crystallin e lattice. Table 1 2 summarizes the properties of the vinylidene halide polymers. The most important trend is observed in the unit cell dimensions of the polymers. For the three polymers in the monoclinic phase and defined by comparable unit cells, it is observed that in increasing the
26 size of the halogen from F to Cl to Br, the a b and c dimensions all increase steadily. This increase reflects the incorporation of the larger halogen in the crystal lattice. 1.2.3 Polytetrafluoroethylene (PTFE) Polytetr afluoroethylene (PTFE), commonly known as Teflon, is represented by the repeat unit (CF 2 CF 2 ) PTFE is synthesized mainly via the suspension polymerization of tetrafluoroethylene with inorganic initiators such as persulfates. 66 Molecular weights in the te ns of millions (g/mol) are estimated for PTFE, although no direct solution measurements can be made, as PTFE is insoluble in all common solvents. 67 The primary structural difference of PTFE relative to all of the previously described halogen containing poly olefins is the symmetry of the repeat unit, which removes tacticity and regiochemistry (head to head or head to tail linkages) as structural factors that must be considered. Additionally, the 92 98% crystallinity observed in untreated PTFE samples reflects a predominantly unbranched chain structure. 68 69 Further, the fluorine atoms are too large to allow the polymer backbone to attain a planar zigzag conformation, and thus the chains lose the flexibility that such a conformation would allow, inducing a high d egree of rigidity into the polymer backbone. 70 At low temperatures the polymer backbone displays a twist of 180 over 13 CF 2 groups, which actually changes to a twist of 180 over 15 CF 2 groups during a transition at 19 C, as will be discussed below. 66 Thermal analysis of PTFE shows a melt transition at 327 C. 71 Several other important thermal transitions are observed, most notably at 19 C, the triclinic crystal structure changes to hexagonal. 67 Whereas, below 19 C, there is almost perfect three dimensional order, the transition at 19 C induces a degree of disorder during the restructuring of the crystalline packing. The cause of this reorganization at 19 C is a slight twisting in the polymer backbone, as described above. Another transition at 30 C further disorders the chain segments, although a hexagonal packing is retained. These two transitions are important because they occur near
27 ambient temperature. 67 The lattice parameters for the highly crystalline triclinic form of PTFE are: a = 9.52 b = 5.59 and c = 17.06 72 PTFE has an excellent thermal stability due to the very strong C F and C C bonds in the poly mer. Very few organic materials possess a thermal stability approaching that of PTFE, which is stable up to nearly 525 C in air. 73 Vacuum degradation studies show an onset at 440 C, which peaks at 540 C. The degradation mechanism of PTFE also differs fro m PVF, for example, in that HF gas is not released. Instead, tetrafluoroethylene is the primary product under vacuum decomposition. 73 Importantly, due to the size of the fluorine atoms and the uniform coverage on the polymer backbone, the C C backbone is protected from chemical attack, conferring excellent chemical stability to PTFE as well. Based on the excellent thermal and chemical stability as well as the excellent mechanical properties, PTFE finds numerous applications. The largest application area is in wiring and electronics, although a number of other applications are found in coatings and seals. 66 A comparison of the properties of the halogen containing hom opolymers leads to several definite conclusions. First, the substitution of a halogen atom for one or more of the hydrogens in the ethylene repeat unit leads to clearly defined changes in the structural properties of the polymer, which has definite consequ ences on the macroscopic properties and strongly influences the applicability of the polymer. Second, it is clear that in such homopolymers within a structure type (e.g. vinyl polymers), increasing the halogen size leads to an expansion of the unit cell (s ee Table 1 1 and Table 1 2) and a consequent change in the crystalline properties of the polymer. Finally, it is observed that increasing the number of halogens per repeating unit leads to a change in the symmetry of the unit cell, as all PVH polymers show an orthorhombic unit cell analogous to PE, but all of the PVDH polymers exhibit a monoclinic unit cell. The reduction in symmetry
28 with increasing halogen content reflects how the incorporation of larger atoms in place of hydrogen disturbs the packing of t he PE backbone. The triclinic packing of PTFE shows a further degeneration in the symmetry of the PE lattice. In all cases such effects on the crystalline properties reflect the incorporation of the halogen into the crystal lattice in such homopolymers. N otice also that the halogen size and content affects the attainable level of crystallinity in these polymers. However, no universal correlation can be drawn based on halogen size and content alone, as such factors as tacticity and defects (e.g. head to hea d linkages) also play a strong role in the ability and extent of the polymers to crystallize. Importantly such homopolymers lead to a limited number of combinations of halogen content and distribution and thus do not allow for the derivation of a complete set of fundamental structure property relationships. For this task, most effort has been focused on ethylene vinyl halide (EVH) copolymers, synthesized via a variety of methods as discussed in the following section. 1.3 Ethylene Vinyl Halide (EVH) Copolyme rs In the previous section, we have examined halogen containing homopolymers for the effect of halogen content, distribution, and size on the thermal and crystalline properties of the polymer. However, the range of halogen contents and distributions in su ch homopolymers, which are almost universally synthesized via radical polymerization from vinyl monomers, is limited. Therefore, it is interesting to examine halogen containing polyolefins synthesized via a number of other methods, which are more appropria te for generating a large range of halogen contents and distributions. In this way, a deeper understanding of the effects of halogen incorporation on the structural and morphological properties of polyolefins can be pursued. The most commonly employed tech niques, to date, are halogenation of PE or analogues, dehalogenation of halogen containing homopolymers, or copolymerization of ethylene with vinyl halide monomers. In the latter method, direct free radical copolymerization of vinyl
29 halides, such as vinyl chloride, and ethylene, it is difficult to control the content of the two monomer units over the full range of compositions. 74 Further, the free radical nature of the polymerization leads to poorly defined structures with high defect contents. 75 76 Direct cop olymerization of vinyl chloride and ethylene using Ziegler Natta or metallocene catalysts is also problematic based on the occurrence of side reactions with the activated chlorine atom on the monomer and the alkylaluminum cocatalyst as well as the tendency of the vinyl chloride monomer to undergo halo elimination after insertion into the metal alkyl bond. 75 77 Based on these limitations, direct copolymerization has not been successfully used to investigate the compositionally dependent bulk properties of halogen containing polyolefins. Here the methods of direct polymer halogenation and dehalogenation will be examined. 1.3.1 Halogenation of Polyethylene Direct halogenation of polyethylene has been extensively used as a technique for developing a set of structure property relationships in halogen containing polyolefins. 78 79 80 81 exhibit a large range of halogen contents and d istributions. The most relevant examples of this class of EVH copolymers are discussed below. 126.96.36.199 Fluorinated PE The surface fluorination of PE has been the subject of numerous studies, largely focusing on how the introduction of fluorine into the su rface layers will modify the surface characteristics of the polymer, such as wetability and permeability. 82 Common methods of surface fluorination are based largely on the treatment of the polymer sample with either fluorine gas or some mixture of fluorine gas in inert gaseous mixtures to yield surface fluorinated PE via a free radical fluorination process. 83 84 Plasma based fluorination schemes are also employed using agents such as CF 4 or SF 6 85 86 In most cases the only changes in properties are confined to t he surface layers
30 of the polymer and the bulk properties remain essentially unchanged. 87 As such, this technique has not been used to generate homogenous samples of fluorinated polymer for study of the bulk structural properties of the material. 188.8.131.52 Ch lorinated PE (CPE) The chlorination of PE is conducted using a variety of methods, such as solution, suspension, solid state, single crystal, thin film, and in the melt. The physical properties of the chlorinated polymers are strongly influenced by the st ructure of the polyethylene sample used as well as the specific chlorination method and conditions employed. 88 Specifically, the rate of chlorination and the polymer structure, as well as the chlorination condition (i.e.; solution, solid state, etc.) influe nce the chlorine content and sequence distribution in the final polymer sample. 78 This complex interplay of factors can lead to an enormous variation in polymer structure and physical properties among chlorinat ed PE samples. Chlorination of PE in solution results in the most random distribution of chlorine throughout the polymer. 81 Solution chlorination specifically refers to the case where polyethylene is complete ly dissolved in a given solvent such that every site on the macromolecules are able to react with chlorine. When the chlorine content is less than 50% by weight, the probability of chlorine substitution on a CHCl unit or neighboring unit is very small base d on the bulky Cl atom, resulting in a random spatial distribution of Cl atoms along the backbone. At greater than 50% by weight, however, geminal chlorination and vicinal chloroination becomes more commonplace, 89 90 with vicinal being favored. 91 The most com monly used solvents are chlorohydrocarbons, such as tetrachloroethane, tetrachloromethane, dichloromethane, or chlorobenzene. Chlorination is effected in a variety of ways, commonly using irradiation, thermal initiation, or radical initiating species in th e presence of chlorine gas. The consequence of this completely random chlorination process is the regular variation in
31 physical properties of a series of CPE samples prepared from a specific PE sample, where the percentage of chlorine introduced into the b ackbone is the primary factor influencing physical properties. Variation in the properties of the initially used PE sample will however introduce a further level of complexity in comparing CPE samples prepared from different PE samples. The variation in thermal properties of solution chlorinated PE reflects the random nature of the chlorination process. Increasing the chlorine content results in a decrease in the flexibility of the polymer backbone, which is reflected by the regular increase in glass tran sition temperature with increasing Cl content. This regular increase also points to the random distribution of Cl atoms in the polymer. The consequence is that CPEs possessing intermediate degrees of chlorination (~15% by weight) show elastic properties, w hile those samples with high degrees of chlorination (>50% by weight) are rigid. 81 90 92 The overall crystallinity and melting point of CPE samples decreases with in creasing chlorine content and above 37 wt% chlorine, crystallinity is totally inhibited. 90 The gradual disappearance of crystallinity with increasing Cl content is explained by the increasing incorporation of C l into the crystalline phase at higher chlorine contents in the random polymers. The incorporation of the Cl atoms into the crystalline phase as defects results in the observed lower melting temperatures, smaller crystallite size, reduced crystallinity, an d decreased degree of crystalline perfection. 93 The distribution of Cl atoms in CPE has been studied by DSC and WAXD. Diffraction data show that the a and b dimensions of the unit cell gradually expand with increasing chlorine content, which is attributed t o the accommodation of the chlorine atoms into the crystalline lattice. While the PE starting material shows an orthorhombic crystal structure, the samples of CPE with the higher chlorine contents show distorted orthorhombic or pseudohexagonal in the extre me of degree of chlorination. 93
32 Closely related to solution chlorination is another solution based technique; suspension chlorination. The main difference between solution and suspension chlorination is that u nder suspension conditions, the PE is not fully dissolved and thus all sites on the polymer backbone are not equally susceptible to attack by chlorine. The consequence is a distinct property set reflecting the non random and often blocky chlorinated struct ure. 93 Suspension chlorination of PE is generally carried out using a water suspension method 94 or using high concentrations of PE in organic solvent at temperatures below the melting point of the polymer. 79 In either case, chlorine gas is introduced into the system and initiation is commonly affected thermally or by irradiation. Chlorination occurs primarily in the amorphous regions 80 of the polymer or at the edge of the crystalline domain. 93 With higher chlorine contents, diffusion of chlorine gas into the crystalline region results in chlorination of the crystalline phase as well. The chlorination of the crystalline phase occurs from the outside edge of the crystal inward and results in the simultaneous conversion of the crystalline to amorphous regions. 93 This two step process is responsible for the blocky nature of the resulting CPE polymers and also leads to the significantly higher chlorine contents in the amorphous regions of the polymer, which display much higher contents of geminal dichloro species relative to the solution ch lorinated PE of the same chlorine content. The thermal behavior of suspension chlorinated PE reflects the non random substitution pattern. The as prepared samples show melting temperatures and thermogram shapes equivalent to that of the parent PE sample on the first scan, but with proportionally decreasing enthalpies of melting ( H f ) relative to the increasing chlorine contents. Subsequent scans show that melting temperatures have shifted to slightly lower temperatures and have become broader with measure d enthalpies proportionally smaller than those measured before the initial melt. This
33 behavior is indicative of the behavior of a block copolymer in which the chlorine rich segments remain amorphous and the ethylene segments are the only ones capable of me lting and recrystallizing. Thus the important factors are the degree of chlorination, which determines how much of the initially crystalline PE domains remain non chlorinated and the thickness of the original crystalline domains. These conclusions are supp orted by WAXD, which shows that the unit cell dimensions, a and b for the suspension chlorinated PE does not change with increasing chlorine contents, indicating that chlorine is essentially excluded from the crystallites. This is in stark contrast to the case of the solution chlorinated PE as described above. Regardless of the chlorine content, samples of suspension chlorinated PE retain the orthorhombic crystal structure of the parent PE, but with decreasing diffraction peak intensity with increasing chl orine content. Based on these results and in comparison with the results from solution chlorination, it is clear that chlorine content and distribution are critical factors affecting the overall degree of crystallinity and the crystalline structure of CPE samples. An extension of suspension chlorination is the chlorination of PE single crystals in suspension. Here large single crystalline PE samples are prepared via careful purification and crystallization, which gives uniform single layer crystals. 95 In this case, chlorination is performed in a suspension of the PE single crystals where the chlorine gas is introduced in the presence of an initiator. Here, chlorination proceeds via a two step mechanism in which rapid chlorination occurs first at the ch ain folds and then is followed by a slower chlorination and subsequent dissolution of chain segments on the edge. The polymer then chlorinates inwards from the edges generating a bimodal polymer composition with a highly soluble chlorinated fraction and an insoluble crystalline fraction with a very low chlorine content. The same is true
34 for the chlorination of PE single crystals in the solid state with either initiation via irradiation or in the absence of irradiation. 96 97 Like suspension chlorination, sol id state chlorination is a process in which PE samples are exposed to chlorinating conditions where differential reactivity exists between different portions of the polymer. Chlorination in the solid state can be performed on PE powders or films in additio n to single crystals as mentioned above. For samples in the solid state chlorine gas is generally passed over the PE sample and initiation is either thermal or radiation based. The characteristics of the chlorination reaction for powdered samples reacted b elow the melting temperature of the PE sample are very similar to those for suspension polymerization, showing high degrees of chlorination in the amorphous regions and chlorination to a lesser extent in the crystalline regions. 78 No change in the unit cell dimensions are observed in the powdery samples, indicating that chlorine is not incorporated to a large extent in the crystalline portions of the polymer samples and relatively little change in the a dimensio n is observed even in the melt crystallized samples, also indicating that even upon recrystallization little chlorine is incorporated into the PE crystal lattice. This behavior is typical for blocky polymers and is similar to the behavior observed in suspe nsion chlorinated PE. 98 The results for thin film samples are also quite similar 99 100 and point to the fundamental consequence of using a heterogeneous chlorination system as opposed to the homogeneous system used in solution chlorination. In the solution ca se, all sites on the polymer chain are equally likely to be chlorinated at early reaction times and at higher chlorination contents, a spatially distributed chlorine pattern is favored, further randomizing the CPE structure. However in the case of heteroge neous chlorination as observed in suspension or in the solid state, differential reactivity exists not only between the amorphous and crystalline regions, but between
35 the different surfaces of the crystal and between the exterior and interior of the crysta lline domains. The consequence for all of the heterogeneous chlorinations is a blocky primary structure as reflected by thermal and structural characterization. 184.108.40.206 Brominated PE The most common method for the bromination of PE is via suspension bromination in the presence of elemental bromine and light. 101 This technique is used for the free radical bromination of either semicrystalline PE or single crystal PE. Common solvents for suspension bromination are bromobenzene, tetrachloroethylene, or car bon tetrachloride. The amount of bromine introduced into the PE sample is a function of the sample type (single crystal or semicrystalline) and the reaction time. Evaluation of the resulting thermal and crystalline properties provides information on the ef fect of bromine content, with respect to the distribution. Suspension bromination of semicrystalline high density polyethylene (HDPE) 102 results in the generation of a blocky rather than a random distribution of bromine atoms along the PE backbone. It is o bserved that bromination occurs only in the amorphous regions of the polymer. Further, the greater the extent the bromination, the greater the decrease in the overall crystallinity of the polymer in melt recrystallized polymer samples. Such behavior leads to the conclusion that the bromine atoms are behaving as a defect, limiting the ability of the polymer to crystallize. Long crystallizable methylene sequences are retained in the polymer after bromination, but their ability to crystallize via a mechanism based on the selection of such long sequences is inhibited as the bromine content of the polymer increases. The melting point of such blocky brominated polymers after melt recrystallization is gradually depressed from the 130 C observed for the pure HDPE to a value of ~124 C in all samples with at least 6% bromine by weight. This leveling off effect is a characteristic of blocky copolymers and specifically characteristic of such samples in which additional bromination beyond 6 wt% does not increase the nu mber of non
36 crystallizable sequences, but simply increases the density of bromine atoms within such sequences, which are formed in the amorphous region of the initial polymer sample. Suspension bromination of single crystal PE produces quite different res ults. 103 In this case, bromination occurs almost exclusively at the fold surface of the crystals. Examination of the orthorhombic unit cell parameters show almost no change for the as brominated samples, as would be expected for the case when bromine atoms a re only added at the fold surface and not within the bulk crystal. Upon melt annealing, both the a and b dimensions (the c dimension the chain direction shows no change) of the unit cell increase steadily with increasing bromine content up to 6.7% brom ine by weight. Beyond 6.7% bromine and up to 19.1% (the highest value measured), the a and b dimensions stay relatively constant. This phenomenon can be explained by the fact that bromination of the single crystal starting material only occurs at the fold surface. After the as brominated sample is melted and it is cooled, the chains reorganize themselves during the recrystallization process, and the incorporation of bromine defects into the crystal is consistent with the increasing a and b dimensions of the unit cell. A steady increase in the a and b dimensions up to about 6.7% bromine is consistent with the presence of an increasing number of defect sites that are available for incorporation into the crystal. Beyond this bromine content, essentially all of the original fold surfaces contain at least one bromine atom and start to be subject to multiple brominations. As such, beyond 6.7% bromine, the a and b dimensions stay essentially constant, as the number of defect sites has leveled off. The changes in H f with increasing bromine content can also be explained. For the as brominated samples, the H f is essentially constant for all samples. A slight decrease of less than 4% in the H f value is observed with 19.1% bromine by weight. In this case an average of >4 bromine atoms are present per fold. As a result, with such a high concentration of bromine
37 present at the fold surfaces, a significant amount of strain is expected to be induced in the crystalline region adjacent to the fold surface, resulting in a d ecrease in H f For the melt recrystallized samples, a steady decrease in H f is observed with increasing bromine content. This steady and large decrease (a 58% decrease is observed for the sample with 19.1% bromine) is consistent with the incorporation of defects in to the crystal lattice. Beyond 6.7 wt% of bromine, the continual decrease in H f reflects the incorporation of larger defects (multiply brominated fold surface segements) rather than the incorporation of additional defect sites. Further, based on this mod el and the results of the XRD and changes in H f discussed above, the changes in T m with increasing bromine content can also be explained. For the as brominated samples, the T m onsets are essentially constant across the entire range of bromination. However for the melt recrystallized samples, the T m onset shows a steady decrease up to the point where one bromine atom per fold surface had been introduced; i.e., up until the point where the number of defect sites remains constant. The incorporation of such d efects into the crystal reduces the level of perfection and simultaneously reduces the T m Beyond this level of bromination, larger decreases in the T m onset reflect the incorporation of larger defects into the crystal, which further distort the structure and decreases the T m 1.3.2 Dehalogenated Homopolymers Another method that has been successfully employed for the synthesis of EVH copolymers is the reductive dehalogenation of halogen containing homopolymers. This method has been employed for the reduct ive dehalogenation of PVC 104 105 106 and PVB. 107 However, the only extensive structural characterizations of the resulting EVH copolymers have focused on EVC generated from PVC. The common synthetic method employed in this case is the reductive dechlorination by r eaction with tri n butyltin hydride. 108 Specifically, PVC is dissolved in THF
38 and in the presence of a catalytic amount of AIBN, a 20% excess of tri n butyltin hydride, relative to the desired level of dechlorination, is reacted with PVC, as seen in Figure 1 4. As a consequence of the homogeneous reaction conditions, a random like distribution of vinyl chloride and ethylene comonomer units is achieved. In this way, a family of copolymers with equivalent chain lengths and polydispersities is achieved by using a single batch of PVC to generate such a family. 108 Also, a family of EVC copolymers with varying comonomer compositions can be achieved, that would not be possible via direct copolymerization of ethylene and v inyl chloride. The random like distribution allows one to study the effect of halogen content and sequence distribution without the limitations of the previously described halogenation methods, which generally selectively halogenated only certain portions of the polymers. In the most complete study of a family of EVC copolymers generated via the reductive dehalogenation of PVC, samples with varying chlorine contents ranging from pure PE (fully reduced PVC) to a sample with 37.3 mol% vinyl chloride were use d. 108 Analysis by WAXD showed that as the chlorine content of the samples increased a progressive change in the crystalline properties of the polymers was observed. First, an expansion of the orthorhombic unit cell of PE was observed, mainly in an increase in the a dimension from 7.48 in pure PE to 8.95 in the EVC sample with 37.3 mol% vinyl chloride (VC). This corresponds to a 20% expansion in the a dimension of the unit cell. Concurrent with the increasing dimension of the unit cell is a gradual change in the crystalline lattice from orthorhombic to pseudohexagonal. Between 13.6 and 21.2 mol% VC units, the change from orthorhombic to pseudohexagonal is observed. Both of these results indicate that chlorine is incorporated into the crystal lattice as a defect. At low concentrations of chlorine, the defect is easily incorporated into the crystal lattice with little
39 disruption of the overall crystalline structure. However, with increasing chlorine content, the presence of a larger amount of defect not only decreases the ordering within the crystal lattice, but also decreases the overall crystallinity on the sample as the defect inhibits crystallization. 74 Therefore, the percent crystallinity of the samples decreases from more than 60% with pure PE to essentially 0% when 40 mol% VC units are present. The thermal properties of these random EVC copolymers also supports the incorporation of VC units into the crystal latti ce as a defect. Here, DSC indicates that the T m of the samples decreases from 128 C with PE to 20 C with the 37.3 mol% VC sample. 1.3.3 Synthesis of EVH Copolymers via Ring Opening Metathesis Polymerization (ROMP) Followed by Hydrogenation Another meth od for synthesizing random EVH copolymers has been recently described. In this case ring opening metathesis polymerization (ROMP) 109 has been employed to achieve random EVC copolymers via the copolymerization of cyclooctene and 5 chlorocyclooctene, followed by hydrogenation as seen in Figure 1 4. By varying the comonomer ratios, vinyl chloride contents from 0 50% were achieved. Analysis via DSC and WAXD was used to examine the crystalline properties of these polymers. As the VC content increased from zero to 13.5 mol%, the orthorhombic unit cell was observed to expand and the reflections in the XRD became broader, both of which indicate that the chlorine atoms are entering the orthorhombic crystal lattice. In fact at 13.5 mol% VC, the unit cell is actually tra nsformed to a distorted orthorhombic or pseudohaxagonal unit cell. With higher VC contents (16, 20.8, and 25 mol%), a true hexagonal unit cell is observed. The crossover in behavior after 13.5 mol% VC is also observed via DSC. For zero to 13.5 mol% VC, the T m gradually decreases, the melting peak becomes broader, and the overall crystallinity decreases with increasing VC content. These three features are consistent with the incorporation of chlorine, as a defect, into the crystal. For the
40 samples with 16, 2 0.8, and 25 mol% VC, the T m shows only the slightest changes between samples and shows an increasing sharpness with increasing chlorine content. The authors postulate that the unusual melting behavior at high VC contents may be a result of regular chlorine placement along the chain. The authors also suggest that such regular placement of chlorine atoms along the polymer chain would nicely explain the observed hexagonal crystal structure at high VC contents, as the hexagonal form is generally not encountered in ethylene copolymers with high comonomer content and the hexagonal form could be expected to accommodate the chlorine atoms more easily than the orthorhombic form. 1.4 Acyclic Diene Metathesis (ADMET) Polymerization as a Route for Deriving Structure Pro perty Relationships in EVH Polymers Considering the above methods for attaining EVH polymers, it is clear that control over halogen content can be achieved in each of these methods, but control over sequence distribution cannot be attained. Therefore, eve n though the properties of EVH polymers with different halogen contents are known to vary, deriving exact relationships between primary structure and physical properties is unattainable because a precise distribution of comonomer units between crystalline and amorphous domains has yet to be determined. 110 Despite the lack of a precise model for understanding the variation in the properties of EVH polymers, there is great practical interest in these materials. 110 S pecifically, the introduction of ethylene units into a PVC chain will serve the role of an internal plasticizer, which can give many advantages over the introduction of an external plasticizer and lead to a new range of materials properties from a single c omponent material. 111 From a more fundamental standpoint, this class of polymers is also of great interest for developing a precise understanding of the relationship between molecular structure and physical properties in semicrystalline polymers in general, where EVH polymers could serve as an all purpose model for this class of polymers. As
41 described above however, no specific catalyst system has been developed to effectively copolymerize ethylene and vinyl halides, leading to the necessity to look toward mo del EVH polymers to establish the fundamental significance of this class of polymer. Such a focus on model polymers will not only help to derive structure property relationships, but will shift focus to the development of effective catalysts for the synthe sis of EVH copolymers. 109 It is difficult however to find suitable model systems that will allow the development of precise structure property relationships 112 and will not be limited by the quality of the synthe tic or preparative method. Ayclic diene metathesis polymerization (ADMET) 113 has be en used to achieve an EVC polymer with both precisely controlled chlorine content and sequence distribution, 114 here such a level of control allows one to derive precise relati onships between polymer primary structure and physical properties in EVC polymers. Further, ADMET allows the investigation of a much greater range of polymers with precisely distributed chlorine atoms, as the linear dienes used in ADMET are free of the structural constraints required for monomers used in ROMP. Presented here is the full characterization of new expanded family of precise and random EVH polymers (Figure 1 4) in which a range of precise halogen conte nts and precise and random sequence distributions are examined in order to generate structure property relationships useful for developing a deeper understanding of the importance of primary structure control in EVH polymers.
42 Table 1 1. Properties of poly(vinyl halide) polymers. Polymer T m (C) X c (%) Unit cell type Unit cell dimensions () Decomposition temperature (C) PE (HDPE) 125 134 62 80 Orthorhombic a = 7.42 b = 4.94 c = 2.55 348 PVF 190 20 60 Orthorhombic a = 8.57 b = 4.95 c = 2 .52 450 PVC 120 260 <10 Orthorhombic a = 10.6 b = 5.4 c = 5.1 200 PVB ------------------Orthorhombic a = 11.0 b = 5.6 c = 5.1 <100 PVI ------------------------------------<100 Table 1 2. Properties of poly(vinylidene halide) polym ers. Polymer T m (C) X c (%) Unit cell type Unit cell dimensions () Decomposition temperature (C) PVDF 172 ( ) 178 ( ) 185 ( ) 50 70 ( ) Orthorhombic ( ) Orthorhomnic ( ) Monoclinic ( ) a = 4.96 ( ) b = 9.64 c = 4.62 a = 8.58 ( ) b = 4.91 c = 2.56 a = 4.96 ( ) b = 9.67 c = 9.20 350 PVDC 198 205 ---------Monoclinic Monocl inic a = 6.73 b = 12.54 c = 4.68 a = 22.54 b = 12.53 c = 4.68 200 PVDB ------------------Monoclinic a = 25.88 b = 13.87 c = 4.77 <100
43 Figure 1 1. S tructure of poly(vinyl halide) p olymers Figure 1 2 Unit cell parameters. (a) Ge neralized unit cell showing axes ( a b c ) and angles ( ). (b) Parameters for specific unit cell types. (c) Orthorhombic unit cell for polyethylene showing unit cell dimensions and superimposed polyethylene chains. (d) Single crystal of polyethylene illustrating the chain folded lamellar structure.
44 Figure 1 3 Structures of poly(vinylidene halide) p olymers Figure 1 4 Routes to EVC p olymers
45 CHAPTER 2 SYNTHESIS AND CRYSTA LLIZATION OF PRECISI ON ADMET POLYOLEFINS CONTAINING HALOGENS 2.1 Introduction Polyolefins are among the most important large volume polymers produced today. 1 Polyethylene (PE) has found use in such diverse areas as packaging, biomaterials, microelectronics, and protective coatings, 115 all of which depend on the tunable semicrystalline morphology of the polymer. 116 Modification of the parent structure through halogen incorporation, as in the case of poly(vinyl chloride) (PVC) or poly(tetrafluorethylene) (PTFE), 117 further extends the range of applications as a result of changes in polymer structure at all levels. 118 119 120 Ethylene vinyl halide (EVH) copolymers also are of interest. Ethylene vinyl chloride (EVC) copolymers offer improved thermal stability relative to PVC. 121 Similarly, ethylene vinyl fluoride (EVF) polymers are of value, such as poly(vinyl fluoride) (PVF), poly(vinylidene fluoride) (PVF 2 ), and poly(ethylene alt tetrafluorethyle ne). 122 Ethylene vinyl bromide copolymers 123 and partially brominated polyethylene 102 are less well known. Synthesizing copolymer variants of common halogenated polyolefins can be challenging due to reactivity rati o issues present during copolymerization 124 125 and due to the ineffectiveness of post polymerization attempts to produce well defined structures. 125 126 Of the various techniques that have been used to synthesize E VH copolymers, the simplest approach is halogenation of PE, a procedure which gives an irregular distribution of halogens along the polymer backbone. Further, the properties of the resulting polymer are highly dependent on the structure of the PE used and the conditions employed for the halogenation. 125 126 Another typical Reproduced with permission fr om Boz, E.; Wagener, K. B.; Ghosal, A.; Fu, R.; Alamo, R. G. Macromolecules 2006 39 4437 4447. Copyright 2006 American Chemical Society.
46 post polymerization method is reductive dehalogenation, 125 where chemical reducing agents such as tributlytin hydride are used to replace halogens with hydrogens, yielding materials that experience the same structural irregularities problematic in all post polymerization strategies. The direct free radi cal copolymerization of ethylene and vinyl halide comonomers yields copolymers containing numerous structural defects and consequent instability. 121 Regardless of monomer ratios, the copolymers contain inordinat ely high vinyl contents as a result of reactivity ratio imbalances, a phenomenon that can only be slightly improved by use of ray induced polymerization. 74 Olefin metathesis chemistry offers an alternative s ynthetic route to such polymers, based on the development of well defined late transition metal catalysts. 127 Ruthenium catalyzed ring opening metathesis polymerization (ROMP) can yield EVC copolymers possessing well defined structures. 121 Acyclic diene metathesis polymerization (ADMET) 128 129 has also been used for the synthesis of a precisely defined EVC polymer. 114 With ADMET, variation in the monomer structure gives access to a broad range of precisely defined polymers, allowing direct correlation of structure property relationships. 113 The ability of ADMET to produce PE structures with precisely placed pendant groups is g enerating a wealth of structure/property data at the moment. As has been known for decades now, the random incorporation of relatively small amounts (<10 mol%) of a structural irregularity in polyethylene (PE) generates copolymers possessing thermal proper ties that adhere to thermodynamic principles of phase transitions in two component systems. 130 131 132 The vast quantity of experimental data that exists unequivocally demonstrates that side groups such as ethyl, propyl, vinyl acetate, styrene and others are n ot incorporated into the copolymer crystal lattice. 133 Their solid liquid transition follows the basis
47 phase. 134 On the other hand, smaller side groups such as methyl, chlorine and oxygen can be partially incorporated into the crystalline lattice. 131 132 A different partitioning of the side group has an important impact on the t hermodynamic and physical properties of these copolymers. Thus, melting temperatures of random copolymers with Cl pendant groups and CH 3 branches are significantly higher than those of copolymers with matched compositions of side groups excluded from the c rystal. 131 The differences reflect the fraction of longer continuous crystallizable sequences present in the former type. The analysis of the thermodynamic behavior of random ethylene copolymers with > ~10 mol% branch points in the main chain is more complex. At this branching level, melting temperatures and degree of crystallinity usually deviate upward from the linear trends observed at the lower branching contents, 130 135 136 and divergences are accentuated for copolymers with side groups that can be accommodated in the crystal. 109 137 In reference to thermodynamic principles, concerns are raised about the actual branching distribution in highly branched systems and how closely the crystallization behavior follows models based on selection of crystallizable sequences. On one hand, at large branching contents the usual temperature rising elution fractionation (TREF), crystal lization fractionation (CRYSTAF) or NMR methods used to probe branching distributions either fail or become overwhelming in complexity. 138 139 On the other hand, the decreased length of continuous methylene sequences may induce a different crystallization mod e, as occasionally speculated 109 and in line with observations of a change in crystallographic packing with increasing comonomer content. 109 137 140 141 Understanding the details of the crystallization behavior for these highly branched systems can only be accomplished via model polymers with well defined microstructures. ADMET polymerization
48 leads to polymers with precis ely defined microstructures and, hence, to excellent models to study the crystallization behavior of highly branched ethylene copolymers. ADMET polymerization followed by hydrogenation produces either the linear polymer chain or ethylene copolymer type b ranched molecules that lack any branching composition distribution. 114 128 architecture c according to the repeated structural unit [(CH 2 ) n CHX] y The unique characteristics of these systems permit the study of ideal models of functionalized polyethylenes with precise ly placed substituents in which n and X can be varied independently. They are models for polyethylenes with controlled hydrogen substitution. We now report the ADMET synthesis and crystallization of a series of precision polyethylene structures containing fluorine, chlorine, and bromine located on each and every 19 th carbon (n = 18) along the polymer backbone. Included in this study is the synthesis of the first bromine containing polyolefin with a precisely defined primary structure, demonstrating that th e weaker carbon bromine bond survives both the metathesis and hydrogenation chemistry employed. Not included is the comparative polymer containing iodine, as this precision polyolefin has eluded our efforts to create it. We continue to investigate syntheti c routes to the iodine analogue copolymer, for it will become important in the final analysis of these halogenated polyolefins. For now, the synthesis and primary structure characterization of the fluorine, chlorine, and bromine containing polyolefins pres ented here delineates the opportunity for a detailed understanding of the effects of halogen incorporation in polyolefins. The effect of the substituent on the crystallographic packing is extracted using IR spectroscopy and wide angle X ray diffraction (W AXD), while the partitioning of the substituent
49 between crystalline and non crystalline regions is inferred from solid state NMR spectroscopy expected to be accomm odated in the all trans polyethylene lattice, the systematic increase of van der Waals radius of the substituents in the series allows quantitative data of the degree to which the orthorhombic polyethylene lattice can tolerate atomic hydrogen substitution. The thermodynamic properties of these model systems will be directly correlated to substituent radius and bond lengths and will establish fundamental grounds for strategies to modify structural and thermal behavior of polyethylenes. 2.2 Results and Discus sion 2.2.1 Synthesis The synthesis of fluorine, chlorine, and bromine substituted EVH polymers with a halogen on each and every 19 th carbon required the preparation of diene monomers, followed by ADMET polymerization and subsequent exhaustive hydrogen ation as shown in Figure 2 1. All three halogen containing monomers were derived from a common ketone precursor 1 142 which was reduced to the alcohol 2 143 prior to halogenation. The fluorine monomer (3) was synthesized in good yield from 2 using diethylamino sulfurtrifluoride (DAST). Chlorination was achieved by tosylation of 2 followed by nucleophilic displacement with LiCl, while the bromine monomer (5) was synthesized directly by action of CBr 4 /PPh 3 on the alcohol 2 Monomers 4 114 and 5 were 127 The fluorine monomer 3 was polymerized in solution (methylene chloride) as this mon omer is a solid as opposed to 4 and 5 which are oils. In a similar manner, the ketone starting monomer (1) These unsaturated ADMET EVH polymers were subjected to exhaustive hydrogenation to generate precision halogenated polyolefins. The nomenclature acronym used herein is UPE19X,
50 where U indicates unsaturation and PE19X indicates a polyethylene backbone with a halogen substituent (X) on every 19 th carbon. In the case of UPE19F and UPE19Cl diimide reduction 144 gave clean conversion of the unsaturated polymer to the fully saturated polymer as evidenced by IR and NMR. Diimide reduction proved ineffective for UPE19Br. In this case it is likely that the presence of nucleophilic species in the h ydrogenation mixture resulted in the displacement of the bromine, a mechanistic event facilitated both by the weaker carbon bromine bond and the better leaving character of the bromine relative to chlorine and fluorine. Successful hydrogenation was achieve d for UPE19Br as well as UPE19O (the polymer product of monomer 1) using 145 Polymer molecular weights were found to range from a M w of 11,300 12,000 g/mol for PE19F and PE19O respectively (GPC vs. PE) to 38,000 49,500 g/mol for PE19Br and PE19Cl respectively (GPC vs. PS) as listed in Table 2 1. 2.2.2 Primary Structure Characterization. The primary structure for these well defined halogenated polyolefins was established using a combination of 1 H NMR, 13 C NMR, TGA (thermogravimetric ana lysis), IR spectroscopy, and elemental analysis. The precise structure of the polymers is supported by NMR, while TGA results prove the composition of the polymer through the clearly observed thermal decomposition and release of exact masses of HX (calcul ated HF = 7.0% and HBr = 23.5%, found HF = 7.3 % and HBr = 22.7%), followed by catastrophic decomposition; as illustrated for PE19F and PE19Br in Figure 2 2 and previously observed for PE19Cl 114 The mass loss i n all cases is in accord with the theoretically calculated value. Figure 2 3 shows the IR spectra for thin films of the three halogenated polymers, data which also serve to support the expected primary structure. Here the absence of a peak at 967 969 cm 1 corresponding to the out of plane C H wag ( w ) of an alkene, 145 146 indicates complete
51 hydrogenation of the polymer backbone. Further, the IR spectra support the presence of each of the expected halogens on the polymer backbone. For PE19F, the sharp peak at 1068 cm 1 protruding from a broader underlying peak is characteristic of the C F stretch. 147 The peaks observed at 611, 660, and 802 cm 1 in PE19Cl are characteristic of C Cl stretching vibrations, 126 148 while for PE19Br, the peak at 612 cm 1 is assigned to the vibrations of the C Br bond. 102 These IR data also provide an insight into the crystalline structure of these p recisely halogenated polymers. In this regard, the significant peaks are those found at ~720 and ~1470 cm 1 which correspond to the vibrational modes of the CH 2 sequences in PE analogues. 126 For PE19F, the doub lets observed at 721 730 and 1463 1472 cm 1 are the same as observed in crystalline PE. The band at ~720 cm 1 corresponds to long trans CH 2 sequences and the band at 730 cm 1 is associated with the rocking vibrations of CH 2 sequences of five or more carbo ns. 126 149 The IR spectra for PE19Cl and PE19Br are clearly different from those for the fluorinated analogue PE19F and pure PE, suggesting a distinct difference in the crystalline packing of these polymers. The Cl and Br precision polymers show a single peak rather than doublets at ~720 and ~1470 cm 1 suggesting that these polymers possess similar crystalline features, yet differ from the fluorine and pure polyethylene counterparts. Both WAXD and solid state 13 C NMR studies of the crystalline structure, presented in the following sections support these results. 2.2.3 Wide Angle X Ray Diffraction (WAXD) X ray diffractograms of halogen containing ADMET samples cooled from the melt at 1 C/min are shown in Figure 2 4 together with the patterns of a matched PE19O (with a C=O group on each and every 19 th carbon) and of a linear polyethylene narrow fraction in the same molecular weight range. It is at first evident that all samples display very sharp diffraction peaks, except for the broader pattern of the brominated material. In contrast, random ethylene
52 copolymers with the same type of branching and with similar, and lower, branching levels are known to display much broader WAXD patterns. 108 109 145 Hence, the sharp diffractograms of Figure 2 4 point to a homopolymer rather than copolymer like crystallization behavior of these polyolefins. It is als o evident that while substitution of the hydrogen for O or F every 19 th carbon does not alter the orthorhombic unit cell packing of the linear chain, substitution with a bulkier atom promotes formation of a different crystallographic phase. Diffraction pea ks found at ~19 and ~22 for Cl and Br containing samples were also found in similar precisely placed methyl branched polymers and indexed as (010) and (100) diffractographic planes of a triclinic cell. 150 Therefore, in reference to this pattern, the diffra ctograms of PE19Cl and PE19Br are assigned to triclinic structures. Compared to the orthorhombic packing displayed by PE and polymers with F and O substitutions, the triclinic pattern is a degeneration in the scale of symmetry supporting previous speculati on 150 that a reduced order is needed to facilitate minimum spatial requirements to accommodate the bulky Cl, Br, or CH 3 groups between adjacent molecules in the crystal. The triclinic pattern differs substantia lly from a single diffraction pattern observed for methyl branched polyethylenes or for ethylene vinyl chlorides with the same content of side groups but randomly distributed. 108 109 140 141 145 In these systems single WAXD peaks were associated with defective orthorhombic or pseudohexagonal structures. Therefore, the progression of WAXD patterns of Figure 2 4 clearly evidence that both, type and distribution of the substituent, impact the chain ordering of branched or substituted polyethylenes at the most fundamental level. The 2 values of the diffraction peaks, together with other characterization data, are listed in Table 2 1 for the PE and ADMET precisely substituted polymers. Also listed are X van der Waals radii and C X bond lengths. Inspection of these data allows evaluation of lattic e
53 expansions in reference to geometric constraints imposed by the solute X in the adjoining matrix. Compared to the unsubstituted polyethylene chain, the shift of 2 (110) and 2 (200) in the PE19F and PE19O patterns to lower angular values, reflects the e xpansion of the orthorhombic lattice due to an increase in both, van der Waals radius and bond length of C F and C=O respectively. Similarly, the increase in the spacing of the triclinic (010) plane from 2 = 22.47 (3.96 ) for PE19Cl to 2 = 21.95 (4.05 ) for PE19Br is also explained by the increased radius and bond length of the Br atom. Therefore, the angular shifts follow expectations from the significant incorporation of side groups in the crystalline structure In addition, sharp diffraction peaks in most patterns point to a high correlation of diffractographic planes and, hence, to highly organized crystalline structures. Given the atactic character of the substitution, 114 it appears that for the sample s analyzed here, the orientation of the substituent with respect to the backbone chain has a negligible effect on the correlation between crystallographic planes. Calculated unit cell dimensions and densities of the orthorhombic lattices are listed in Tabl e 2 2. Pertinent data for dimensions of the triclinic lattices could not be obtained since they would require observation of at least five independent WAXD reflections. 150 The diffractograms of Figure 2 4 also indicate that there is a significant amorphous region in these systems that increase with the bulkiness of the substituent. To estimate the degree of crystallinity from the diffractograms, the WAXD patterns obtained at 150 C were scaled and subtracted fro m the room temperature patterns. Crystallinity levels obtained are listed in Table 2 2 1 and range from 83% for PE to ~ 40% for PE19Br. In comparison, random ethylene copolymers with the same level of methyl branching or chlorine pendant groups (~5.3 mol%) display much lower levels (~25%) and less organized crystallinities. 74 109 128 One can then conclude that the initial selection of l ong methylene sequences, present in the random system,
54 over sequences containing for example Cl or Br substituents, imposes additional constraints in the topology of the remaining melt for gathering additional sequences with the required length to propagat e crystallization. More specifically, in the random systems, the crystallization is primarily driven by the selection of continuous crystallizable sequences longer than a critical value, while shorter sequences remain uncrystallized. Clearly, the lack of branching distribution in ADMET polyethylenes with side groups on every 19 th carbon, invokes a homopolymer like crystallization with a driving force led by the accommodation of the side group in an all trans backbone packing conformation. This crystallizat ion mode explains the sharp diffractograms. Thus, while the melting behavior of the random samples is invariably broad, precisely substituted polymers melt sharply as clearly indicated by the melting thermograms of PE21CH 3 and the randomly distributed coun terpart (with matched methyl content) in Figure 8 of reference 145 The change in crystallographic packing of random vs. precisely placed methyl branched ADMET systems is also clear evidence of their different crystallization behavior. 145 The fact that precisely substituted ADMET polymers display relatively high levels of crystallinity suggest a crystalline state built on the basis of substitutional solid solutions. 151 In packing backbone sequences in all trans conformation, substitution of a H for O, F, Cl or Br on each and every 19 th der Waals radius. As seen in Table 2 1, the orthorhombic lattice is preserved in this series up to a radius of ~ 1.6 while bulkier atoms cause large lattice distortions to the point that correlated symmetry between crystallographic planes is only found within a different phase with significantly larger dimens ions. 152 In reference to the hydrogen, the discontinuity of isomorphic structures in this series occurs at a difference in van der Waals radius of ~ 30%. This is larger than the ~ 15% or lower difference usually quoted for formation of isomorphic metallic s olid
55 solutions, 151 but not unexpected, taking into account the chain connectivity of polymer molecules and the weaker covalent bonding. 2.2.4 Melting and Crystallization Behavior A further insight into the uni que crystallization mode of this family of precisely substituted polyethylenes is given by their crystallization and melting behaviors as measured by differential scanning calorimetry (DSC) shown in Figure 2 5. The sharpness of the crystallization and mel ting traces are typical of the behavior of low molecular mass homopolymers, and they contrast with much broader endotherms displayed by random ethylene copolymers of similar branching composition. 74 131 145 Thus, thermodynamic features also point to a mechanism for ordering the precisely branched ADMET polymers that differs from a partitioning of sequences along the cry stallization process, which is typical of the random behavior. 132 135 crystalliza ble short sequences; this feature leads to an inequality in partitioning of the substituent between the crystalline and non crystalline regions. As a consequence, the composition of the melt coexisting with crystallites changes continuously during melting, leading to the broad endotherms usually observed in the random systems. 130 135 153 The peak melting temperatures decrease dramatically and proportionally to the v an der Waals radius in the series of halogenated polymers, as seen in Figure 2 6, projecting a value of ~ 30 C for the melting temperature of the iodine polymer with the same substituent pattern. We observe that for similar van der Waals radii, shorter bo nd lengths increase the melting temperature as evidenced by the ~ 7 degrees higher melting of PE19O compared to PE19F. Similarly, the melting temperature of PE19CH 3 128 is higher than predicted from van der Waal s radii arguments; due to a much shorter C C bond length. The high heat of fusion of PE19F
56 (Table 2 1) is associated with a high electronegativity of the F atom and its likelihood of increasing intermolecular secondary bonding compared to the other members of the series. Clearly, the variation in crystallization and melting temperatures in this series is indicative of the degree to which each substitution perturbs the symmetry of the neighboring carbons in the lattice. Accommodation of the larger atoms must by necessity, distort the all trans ordering of vicinal intra and intermolecular carbons resulting in a more defective structure and, as a consequence, in significantly altered thermodynamic data. 2.2.5 Solid state NMR Crystallographic and thermal prop erties of the series of precision substituted polyethylenes, such as the change and expansion of the unit cell and over 75 C decrease in melting temperature with increasing van der Waals radius, were interpreted as the result from the need to pack in crys talline arrays continuous molecular segments that include the side groups. In this section, the inclusion of halogens in the crystalline regions is probed by direct solid state NMR analysis of resonances belonging to or associated with the side group in or dered and disordered environments. Of the ADMET polymers investigated, PE19F is the best suited candidate to reveal differences between crystalline and non crystalline side groups. In this sample, direct polarization solid state 19 F NMR provides informati on on all fluorine atoms at high sensitivity while information via 13 C NMR of the solid samples relies only on ~1% natural abundance of the 13 C isotope. The directly polarized 19 F spectrum of PE19F obtained under high power 1 H decoupling (DD) and magic an gle spinning (MAS) of 14,000 Hz is given in Figure 2 7a. The spectrum was obtained with a recycle time of 20 s to ensure complete F magnetization recovery. Paralleling the expected incorporation of a large fraction of F atoms in the crystalline regions, t wo distinctive resonances are observed associated with fluorine atoms in crystalline ( 179.4 ppm) and non
57 crystalline ( 181.9) environments. The integrated intensity associated with each resonance was extracted from a fit of the total intensity with two Lo renztian peaks, which led to a crystallinity of 56% in very good agreement with the 57% value obtained by WAXD for the same sample. Such a good correspondence is expected if PE19F crystallizes as a homopolymer with no discrimination in the partition of th e F side groups between the different phases of the crystalline structure. Contrasting with a clear difference in chemical shift between ordered and disordered F, the spin lattice relaxation times (T 1 F ) associated with both regions are very similar, 3.85 s and 3.36 s respectively, denoting the lack of specific orientation of the F with respect to the backbone chain and the defective nature of this atom in the crystalline lattice. Although the solid state 19 F NMR spectrum of Figure 2 7a evidences clearly th e presence of F atoms in two different conformational environments, it does not provide direct quantitative information on the content of F in each region. This information is feasible via solid state 13 C NMR by the correlation between CH and CH 2 carbons pertaining to crystalline and non crystalline environments. Moreover, the 13 C DD/MAS spectrum of PE19F was not recorded as the lower sensitivity of 13 C observation added to our inability to decouple both 1 H and 19 F simultaneously will undoubtedly result i n a broad featureless CH resonance. Solid state DD/MAS 13 C NMR spectra were obtained for the other two halogen substituted polyethylenes of the series and are given in Figures 2 7b and 2 7c for PE19Cl and PE19Br respectively. Prior to acquisition of ful ly relaxed spectra, carbon spin lattice relaxation times for CH, CH 2 and crystalline CH 2 groups in both polymers were estimated as 1.5 s, 2 s and 25 s respectively. Accordingly, fully relaxed single pulse spectra of >2600 transients were collected at roo m temperature under MAS and high power 1 H decoupling using recycle times of 120 s. Small peaks observed at 15.3 ppm are associated with spinning side bands. The main resonance
58 centered at 34.1 ppm corresponds to crystalline CH 2 in the all trans conformatio n and is shifted 1.3 ppm downfield in reference to the crystalline orthorhombic CH 2 resonance of linear polyethylene, which is centered at 32.8 ppm under the same field and experimental conditions. This shift points to a crystallization pattern for PE19Cl and PE19Br that differs from the orthorhombic symmetry as previously observed by WAXD. A correspondence with similar downfield shifts observed in solid n alkanes that pack in a triclinic phase as compared to the n alkane orthorhombic symmetry, 154 serves as additional indirect evidence of the triclinic packing of these polymers. By analogy to the spectrum of solid PE, the resonance at 31.5 ppm is assigned to non crystalline CH 2 carbons and by comparison to the spectrum recorded in solution, peaks centered at 41.2 and 2 and CH carbons of PE19Cl and PE19Br 2 is mostly hidden beneath the 31.5 ppm resonance of the amorphous CH 2 carbons. As previously indicated, the peak correspon ding to the CHX carbon is ideal for monitoring the partitioning of Cl and Br atoms between crystalline and non crystalline regions. For this purpose the information from the CHX resonance is useful when both phases contain a sufficiently large number of CH X groups to overcome the low sensitivity nature of 13 C detection, and when crystalline and non crystalline phases display a resolved difference in their CHX chemical shift. An additional constraint to acquire spectra with high signal/noise ratio in our sa mples was the small quantities of ADMET polymers available for these experiments (~35mg). The spectral region associated with the CHX resonance is expanded in the inset of figures 7b and 7c. Both consist of two largely overlapping resonances centered at 66.9 and ~64 ppm for PE19Cl and at 62 and 59 ppm for PE19B r. In consonance with the resonances observed in PE19F in Figure 2 7a, the CHX observed resonances indicate that at least two populations of
59 methine carbons with significantly different conformatio ns are present in crystalline PE19X polymers. Small protrusions in the downfield wings of the crystalline CHX resonance are attributed to a high noise level from the small quantities sampled. The contribution from the interface will be buried within the c rystalline and non crystalline CHX resonances. Owing to the fact that the difference in chemical shift and the ratio of peak heights of these two resonances are very similar to the values of crystalline (34.1 ppm) and non crystalline (31.5 ppm) CH 2 peaks, both CHX resonances are assigned to crystalline methine (66.9 ppm, 62 ppm) and non crystalline methine (64 ppm, 59 ppm) carbons. The observed chemical shifts for crystalline and non crystalline CHCl in Figure 2 7b are identical to the values reported by To nelli et al. in a study of the structure of ethylene vinyl chloride copolymers using CP/MAS techniques. 108 From the integrated intensities of the combined (crystalline and non crystalline) CHX resonances and t he intensity of CH 2 resonance (~41.0 ppm) relative to the CH 2 resonances (34.1 and 31.5) contents of chlorine and bromine of 5.4 0.3 mol% and 5.3 0.2 mol% are calculated, respectively. These values match the content of halogen in the molecule from the primary structure (5.3 mol%) and evidence the quantitative nature of the DD/MAS spectrum. In other words, all types of carbons of PE19Cl and PE19Br were relaxed after the 120s delay. Since both CH 2 and CHX carbons have very similar distributions in am orphous and crystalline environments, from the equivalence in the corresponding intensity ratios, we conclude that chlorine and bromine groups must be uniformly distributed between the phases. Upon crystallization there is no discrimination for the partit ions of CHX units between crystalline and non crystalline regions. In contrast, the DD/MAS spectrum of a randomly distributed ethylene vinyl chloride with similar chlorine content displayed much broader resonances and no
60 distinction between crystalline an d amorphous CHCl carbons. 108 In the random copolymer at least 20% of the chlorines were estimated to be included in the crystal. The distribution of side groups among the different phases of the semicrystallin e structure was also probed by additional cross polarized (CP) under MAS NMR experiments that isolate the spectra of the crystalline regions. 155 156 Resonances in the crystalline spectra that are associated with the methine carbon or carbons adjacent to the m ethine group allow direct quantitative data on the content of these groups in the crystalline regions and, hence, on the partitioning of the side groups. As mentioned in the experimental part, the CP method used here for isolating the crystal spectra is b ased on differences in T 1 H of crystalline and non crystalline regions. These values were estimated for the more intense ~34 ppm peak and are listed in Table 2 3 for the linear and three halogenated polymers. All the samples display a fast relaxing compo nent, 2 ms < T 1 H < 4 ms, and a slower one (T 1 H > 6 ms), associated with the non crystalline and more rigid crystalline phases respectively. Owing to a modest contrast in crystalline and non crystalline proton mobility for PE19Cl and PE19Br in the freque ncy level at which T 1 H is active, a 5 ms 1 H spin locking filter was applied for these two polymers prior to CP, and a longer one (7 ms) was used to filter most of the non crystalline regions of PE19F. Consequently, crystalline spectra would be more stron gly weighted after the filter. In addition to this filter, to further null or decrease the contribution of the non crystalline carbons, a fraction of the unfiltered CP spectra (shown in Figure 2 8a ) is subtracted from the spectrum generated with spin lock ing. 155 156 The scaling factor for subtraction is chosen as the largest number that leaves no regions with negative intensities in the spectral region between 10 a nd 120 ppm. The resulting crystalline spectra are shown in Figure 2 8b and t he corresponding observed chemical shifts for crystalline carbons in the halogenated series are listed in Table 2 4
61 Unfiltered CP MAS spectra, shown in Figure 2 8a reveal an incre ase of the intensity of the amorphous CH 2 resonance (~31 ppm) with the size of the substituent, thus reflecting a decrease of crystallinity from PE19F to PE19Br in the series, in consonance with the WAXD data. Also of interest is the chemical shift corresp onding to crystalline CH 2 of PE19F, observed at 33.1 ppm, a value that is very close to the all trans orthorhombic linear polyethylene. This correspondence in chemical shifts confirms that PE19F maintains the orthorhombic packing of the PE chain. Resonan ces of carbons associated with the substituents are still observed in the filtered crystal spectra (Figure 2 8b) even after subtraction of most of the amorphou s components. 2 2 (observed at 28.6 ppm for PE19F, at ~31.0 ppm for PE19Cl and buried within the main CH 2 resonance in the spectrum of PE19Br). Observation of these resonances confirms that all th e types of halogen side groups are incorporated in the crystalline regions of these systems. This assertion is at variance with recent predictions from molecular dynamics (MD) simulations of the crystallization process of precisely placed methyl and chlor ine side groups, 157 yet agrees with more abundant qualitative data, estimated from WAXD and DSC, for the random systems 103 158 159 160 and precisely placed methyl branched polymers 150 which indicated that all types of side groups of interest here are incorporated in the crystal lattice. 2 resonances and the resonance associated with the CHX in the crystalline spectra allow us to integrate these lines against the integrals of the main peaks and, thus, deduce the concentration of F, Cl, and Br groups in the crystal. These data are also listed in Table 2 4. Accounting for the larger experimental error associated with integrat ion of broad CHX resonances, the values obtained for the concentration of halogen in the crystal from any of the resonances associated with the side groups are very similar to the
62 concentration of halogen in the chain (5.3 mol%). Therefore, analyses of CP MAS spectra confirm the results obtained from the DD/MAS spectrum of PE19F, PE19Cl and PE19Br. They also confirm the uniform distribution of the halogen in the semicrystalline structure and thus, the homopolymer crystallization behavior of these systems. 2.3 Conclusions The synthesis and unique crystallization behavior of a series of precisely substituted polyolefins, with halogens (F, Cl, Br) or ketone groups on each and every 19 th carbon, has been described, with synthesis of the precise, bromine contai ning polymer (PE19Br) proving to be most challenging. Difficulties in isolating PE19Br were overcome via use of mild hydrogenation and the narrow melting and crysta llization peaks found by DSC for these polymers conform with a homopolymer like crystallization. These observations contrast with much broader diffraction and melting peaks observed in ethylene copolymers with matched concentrations of randomly distribute d side groups. In reference to the random copolymers, precisely substituted ADMET polymers display relatively high levels of crystallinity, as measured by WAXD, and a crystalline state built on the basis of substitutional solid solutions. The accommodation of O, F, Cl or Br preserved up to a radius of ~1.6 (F, O substituents). Accommodation of bulkier substituents (Cl, Br) degenerates the correlated symmetry to a triclinic lattice with significantly larger dimensions. Direct solid state 19 F and 13 C NMR investigations of the crystalline and non crystalline regions evidence a uniform partitioning of all of these substituents between the different phases. While crystallization of the random type is led by the selection of long crystallizable sequences, in these precision polyolefins it is governed by the accommodatio n of
63 the side groups into the crystal. As such, both type and distribution of the substituent impact the chain ordering of branched and substituted polyethylenes at the most fundamental level. 2.4 Experimental Section Chemicals. Chemicals were purchased fr om the Aldrich Chemical company and used as benzylidine ruthenium (IV)dichloride, was purchased from Strem Chemical and stored in an Argon filled dry box prio 3 ) 3 was purchased from Strem. Methylene chloride and o xylene were distilled over CaH 2 PE (M n = 13.1 kg/mol, PDI = 1.26) was obtained from Instrumentation. So lution 1 H NMR (300 MHz) and 13 C NMR (75 MHz) spectra were recorded on a Varian Associates Gemini 300, VXR 300 or Mercury 300 spectrometer. All chemical shifts for 1 H and 13 C NMR were referenced to residual signals from CDCl 3 ( 1 H = 7.27 ppm and 13 C = 77.23 ppm) with an internal reference TMS 0.03% v/v. H igh resolution mass spectral (HRMS) data were obtained on a Finnegan 4500 gas chromatograph/mass spectrometer using the electron ionization (EI) mode. Elemental analyses were carried out by Atlantic Microlabs Inc., Norcross, GA. The GPC measurements for samples in THF were taken on a Waters GPCV 2K instrument Samples were run with HPLC grade THF at 40 0 C on Water StyragelHR 5E columns relative to polystyrene standards Polymer molecular weights reported vers us polyethylene standards were measured using a Waters Associates 150C high temperature gel permeation chromatograph equipped with three Polymer Laboratories mixed bed Type B columns and an internal DRI detector. The mobile phase was BHT inhibited 1,2,4 t richlorobenzene (135 C, flow rate 1.0 mL / minute, typical sample concentration 2 mg / mL).
64 IR data was obtained using a Perkin Elmer Spectrum One FTIR outfitted with a LiTaO 3 detector. Measurements were automatically corrected for water and carbon dioxi de. Thermogravimetric analysis (TGA) data was obtained with a Perkin Elmer 7 series thermal analysis system. The TGA samples (2 5 mg) were heated from 10 C to 800 C at 10 C/min. Melting and crystallizations were obtained at 10 C/min in a Perkin Elmer differential scanning calorimeter DSC 7 with Pyris software under nitrogen flow and calibrated with indium. WAXD diffractograms were collected at room temperature on samples crystallized from the melt at ~ 1 C/min using a slit collimated Siemens D 500 d between 5 and 40 with a step size of 0.02. The instrument was calibrated for d spacing with a standard polished piece of polycrystalline quartz, and the film thickness was offset using shims. The diffractogram of molten polye thylene, used to estimate the degree of crystallinity, was collected at 150 C on a Siemens D500 with an attached Anton Paar HTK high temperature head. DD/MAS solid state 19 F NMR spectra were obtained in a Bruker DRX 600 MHz spectrometer operating at 600.13 and 564.68 MHz for 1 H and 19 F, respectively, under MAS frequency of 14,000 Hz and a decoupling power for 1 H of 71 KHz. Spin lattice relaxation times were obtained via inversion recovery using 19 F 90 and 180 degree pulse length of 2 and 4 s, respectively. Chemical shifts were quoted with respect to the 19 F isotropic signal of flufenamic acid (COOH C 6 H 4 NH C 6 H 4 CF 3 ) at 61.5 ppm as external reference. 13 C NMR spectra of the solid polymers were obtained in a Bruker DMX300 spectrometer operating at 75.5 MHz for 13 C and at 300.2 MHz for 1 H. The experiments for both types of solid state NMR were conducted at room temperature using a Bruker solid state probe for 4 mm rotors under a MAS frequency of 4000 Hz. The small quantities (< 100 mg) available for these p olymers were placed in the center of the zirconium rotors and the empty spaces filled with fine
65 paper tissue or teflon tape. The nutation frequencies associated with the 13 C and 1 H radio frequency fields in the CP experiments were 62 kHz for both isotopes. The 1 H nutation frequency applied for decoupling was 83 kHz. Two different types of 13 C NMR spectra were collected under high power 1 H decoupling, single pulse spectra with recycle delays sufficiently long to fully recover the magnetization of carbons in amorphous and crystalline environments and 1 H 13 C cross polarization combined with MAS (CP MAS) spectra. Contact times for cross polarization were 0.5 0.7 ms. Fixing the CP time and varying a proton spin locking pulse length, prior to cross polarizati on, enables estimation of rotating frame relaxation times, T 1 H for the protons. Values of T 1 H for crystalline and amorphous regions, usually very different, can be used to filter the amorphous phase prior cross polarization. The procedures have been det ailed in previous works. 155 156 Chemical shifts were measured with respect to tetramethylsilane at 0 ppm reference. All peak fit analyses were carried out using the software GRAMS from Galactic Corp. Synthesis. Heneicosa 1,20 dien 11 one (1) was synthesized using a published procedure. 142 The ketone (1) was then reduced to heneicosa 1,20 dien 11 ol (2) using a method similar to that reported. 143 The 1 H NMR and 13 C NMR were in agreement with spectra found previously and the purity of the compounds was supported by ele mental analysis. 11 Fluoro heneicosa 1,20 diene (3). A solution of 5 g (0.031 mol) diethylaminosulfur trifluoride (DAST) in 25 mL CH 2 Cl 2 was cooled to 78 o C and a solution of 5 g (0.016) 2 in 25 mL CH 2 Cl 2 and dry pyridine (2.5mL) were added dropwise. The mixture was stirred at this temperature for 2 hr and then warmed to room temperature and stirred overnight. At this time water was added and the organic phase was extracted with CH 2 Cl 2 and then dried with Na 2 SO 4 and then the solvent was removed to give a w hite solid which was purified by chromatography
66 using 97:3 hexanes:ethyl acetate to give 2.85 g (57%) of the product. 1 H NMR (300 MHz CDCl 3 ) 5.82 (m, 2H), 4.97 (m, 4H), 4.54 4.38 (dp, 1H), 2.05 (q, 4H), 1.7 1.2 (bm, 28H). 13 C NMR (75 MHz, CDCl 3 ) 139.44, 114.35, 64.61, 38.74, 34.03, 29.68, 29.63, 29.39, 29.32, 29.14, 26.71. HRMS calcd for C 21 H 39 F (M + ), 310.3036; found 310.3041. Anal. Calcd f or C 21 H 39 F: C, 81.22; H 12.66; F, 6.12. Found: C, 81.21; H, 12.76; F, 6.03. 11 Chloro heneicosa 1,20 diene (4). To 25 mL of dry pyridine at 0 o C was added 4.5g (2.4x10 2 mol) of tosyl chloride and 6.2 g (2.0x10 2 mol) of heneicosa 1,20 dien 11 ol under arg on. The mixture was stirred overnight. After filtration, 200 mL ether was added into the mixture. The mixture was washed with 1M HCl (50 mL x3), washed with water (50 mL x3), dried over MgSO 4 and the solvent was removed under vacuum to yield a crude yello w oil. Purification by chromatography on silica with ether hexane (15:85) as eluent gave the tosylate which was carried on to the next step. To 150 mL of dry acetone was added 6.8 g (0.16 mol) of LiCl and 5.0 g (0.016 mol) of the 11 tosyl heptadeca 1,20 di ene under argon. After refluxing for 5 days, the mixture was cooled to room temperature. At this time water was added and the solution was extracted with ether (x3). The organic layer was then washed saturated NaHCO 3 followed by brine and then dried with M gSO 4 After filtration the solvent was removed under vacuum to yield clear oil. The clear oil was purified by flash column chromatography using hexane to yield the desired product as a clear oil (3.7 g, 57%). 1 H NMR (300 MHz CDCl 3 ) 5.79 (m, 2H), 4.79 (m, 4H), 3.86 (m, 1H), 2.02 (q, 4H), 1.69 (m, 4H), 1.50 1.27 (br, 24H). 13 C NMR (75 MHz, CDCl 3 ) 139.13, 114.12, 64.29, 38.50, 33.80, 29.45, 29.40, 29.16, 29.09, 28.90, 26.48. Anal. Calcd for C 21 H 39 Cl: C, 77.14; H 12.02; Cl, 10.84. Found: C, 77.15; H, 12.11; C l, 10.57.
67 11 Bromo heneicosa 1,20 diene (5). A solution of heneicosa 1,20 dien 11 ol (2) (5 g, 1.6x10 2 mol) and CBr 4 (6.3g, 1.9x10 2 mol) in CH 2 Cl 2 (25 mL) was prepared in a 500 mL flask and cooled to 0 o C. Triphenyl phosphine (6.2g, 2.4x10 2 mol) was add ed slowly with stirring. Upon addition of the phosphine, the colorless solution turned a pale brown color and was stirred for an additional 4 h at room temperature. The crude product was concentrated under reduced pressure and phosphine oxide was recrysta llized from diethyl ether at 20 C. The solution was filtered and the product was purified by flash chromatography using hexane as the eluent to give 5.2 g (88%) of the product as an oil. 1 H NMR (300 MHz CDCl 3 ) 5.80 (m, 2H), 4.95 (m, 4H), 4.01 (m, 1H), 2.05 (q, 4H), 1.78 (m, 4H), 1.50 1.27 (br, 24H). 13 C NMR (75 MHz, CDCl 3 ) 139.14, 114.11, 58.87, 39.16, 33.77, 29.41, 29.38, 29.07, 29.03, 28.89, 27.55. HRMS calcd for C 21 H 39 Br (M + ), 370.2235; found, 370.2229. A nal. Calcd for C 21 H 39 Br: C, 67.90; H 10.58; Br, 21.51. Found: C, 68.01; H, 10.57; Br, 21.21. General procedure for bulk polymerization. catalyst were combined in a ratio of 500:1 under argon atmosphere. The polymerizat ion was conducted at 35 40 C under vacuum with stirring for 5 days. The reaction was then stopped and 5 mL of toluene was added to dissolve the polymer with stirring. The reaction was allowed to cool to room temperature. The polymers were then precipitate d by dripping the toluene solution into cold acidic methanol. They were then isolated by filtration and dried. General procedure for solution polymerization. catalyst (500:1 ratio) were dissolved in CH 2 Cl 2 for polymeriz ation of PE19F and in toluene for PE19O under argon and stirred at 35 40 o C for 5 days. The same procedure as described above was used to isolate the polymer.
68 Polymerization of 11 Fluoro heneicosa 1,20 diene (UPE19F). Synthesized by the solution method as above using 1.0 g (3.2x10 3 mol) 3, 5.3x10 3 g (6.4x10 6 generation catalyst. Analytical yield. 1 H NMR (300 MHz CDCl 3 ) 5.80 (b, 0.28H), (m, 2H), 4.95 (b, 0.70 H), 4.55 4.35 (dp, 1H), 1.95 (m, 4H), 1.65 1.15 (b, 30 H). 13 C NMR (75 MHz, CDCl 3 ) 130.64, 130.17, 95.87, 93.65, 35.64, 35.36, 32.85, 29.94, 29.81, 29.80, 29.70, 29.42, 27.51, 25.46, 25.40. Polymerization of 11 Chloro heneicosa 1,20 diene (UPE19Cl). Synthesized by the bulk method as above using 1.0 g (3.1x10 3 mol) 4, 5.0x 10 3 g (6.1x10 6 catalyst. Analytical yield. 1 H NMR (300 MHz CDCl 3 ) 5.80 (b, 0.02H), (m, 2H), 4.95 (b, 0.08 H), 3.85 (p, 1H), 1.95 (m, 4H), 1.65 (m, 4H), 1.55 1.15 (b, 24 H). 13 C NMR (75 MHz, CDCl 3 ) 130.57, 130.10, 64.56, 38.77, 32.81, 29.96, 29.87, 29.72, 29.66, 29.50, 29.43, 29.36, 27.44, 26.74. Polymerization of 11 Bromo heneicosa 1,20 diene (UPE19Br). Synthesized by the bulk method as above using 1.0 g (2.7x10 3 mol) 5, 4.4x10 3 g (5.4x10 6 neration catalyst. Analytical yield. 1 H NMR (300 MHz CDCl 3 ) 5.80 (b, 0.05H), (m, 2H), 4.95 (b, 0.14 H), 4.02 (p, 1H), 1.95 (m, 4H), 1.75 (m, 4H), 1.6 1.2 (b, 24 H). 13 C NMR (75 MHz, CDCl 3 ) 130.75, 130.29, 59.41, 39.62, 33.01, 30.16, 30.05, 29.89, 29.85, 29.69, 29.55, 29.49, 28.01, 27.63. General Procedur e for hydrogenation. The polymers containing F or Cl halogens (UPE19F and UPE19Cl) were then hydrogenated using a modified version of the method described by Hahn 144 by dissolving in dry o xylene under argon an d adding 3.3 equivalents of p toluenesulfonyl hydrazide (TSH) and 4 equivalents of tri n propyl amine (TPA). The solutions were refluxed for 9 hours and then cooled to room temperature. The hydrogenated polymer was
69 precipitated into ice cold methanol and i solated by filtration. The dried polymer was then redissolved in toluene and re precipitated by dipping into ice cold acidic methanol. A white solid was collected by filtration and the polymers were isolated in quantitative yield. The polymers containing Br halogen (UPE19Br) and ketone (UPE19O) were hydrogenated using a 150 mL Parr high pressure reaction vessel equipped with a glass liner and Teflon stirbar nit rogen blanket. Finally, 20 mL of toluene were added. The vessel was sealed and attached to a grade 5 hydrogen tank and purged with hydrogen several times. The bomb was charged with 500 psi of H 2 and stirred for 5 days at 80 C. The hydrogenated polymer was dissolved in toluene, and precipitated into methanol. The polymer was then filtered and dried under reduced pressure. PE19F. Hydrogenation was performed as above. 1 H NMR (300 MHz, toluene d 8 ) 4.43 4.27 ( dm, 1H), 1.7 1.2 (bm, 36H). 13 C NMR (75 MHz, toluene d 8 ) 95.21, 92.96, 36.08, 35.79, 30.30, 30.26, 30.19, 30.16, 25.81, 25.75. M w (GPC vs. PE) = 11,300 g/mol. PDI = ( M w / M n ) = 1.52. Anal. Found: C, 77.62; H, 14.16; F, 5.77. PE19Cl. Hydrogena tion was performed as above. 1 H NMR (300 MHz, CDCl 3 ) 3.86 (p, 1H), 1.68 (m, 4H), 1.55 1.15 (b,32H). 13 C NMR (75 MHz, CDCl 3 ) 64.56, 38.75, 29.92, 29.88, 29.82, 29.75, 29.43, 26.73. M w (GPC vs. PS) = 49,500 g/mol. PDI = ( M w / M n ) = 2.22. Anal. Found: C, 74.98; H, 13.87; Cl, 10.66. A lower molecular weight sample used in the DD/MAS solid state 13 C NMR: M n ( 1 H NMR) = 12,000 g/mol. PE19Br. Hydrogenation was performed as above. 1 H NMR (300 MHz CDCl 3 ) 4.03 (p, 1H), 1.79 (m, 4H), 1.6 1.2 (bm, 32H). 13 C NMR (75 MHz,CDCl 3 ) 59.26, 39.42, 29.93, 29.88, 29.82, 29.74, 29.32, 27.82. M w (GPC vs. PS) = 38,100 g/mol. PDI = ( M w / M n ) = 1.72. Anal. Found: C, 60.58; H, 11.43; Br, 19.54.
70 PE19O. Hydrogenation was performed as described above and the analytical characteriz ation was consistent with that previously reported in the literature. 161
71 Table 2 1. Crystalline data of precision ADMET polyolefins (PE19X) with general structure [(CH 2 ) 18 CHX] y Melting and crystallization temperatures and heat of fusions correspond to sa mples crystallized and melted at 10 C/min. Sample Atom (a) (X) Mn (b) 10 3 (g/mol) Mw/Mn Packing cell 2 (degrees) Tm ( C) Tc ( C) H (J/g) Xc (c) (%) vdW (d) () C X (e) () 110 200 100 010 PE H 13.1 1.26 Orthorh 21.64 24.02 133.0 115 238 83 1.2 1.09 PE19O O 3.4 3.55 Orthorh 2 1.43 23.28 134.7 120 106 60 1.52 1.2 PE19F F 7.5 1.52 Orthorh 21.23 23.60 127.5 113 207 57 1.47 1.35 PE19Cl Cl 22.3 # 2.22 Triclinic 19.10 22.47 72.7 63 105 50 1.75 1.78 PE19Br Br 22.2 # 1.72 Triclinic 19.20 21.95 61.5 43 55 40 1.85 1.95 PE19C H 3 145 CH 3 11.3 1.90 Triclinic 18.75 (f) 21.75 (f) 57 51 96 nd 2.0 1.54 (a) Every 19 th carbon. (b) Measured by GPC vs. PE and # GPC vs. PS.(c) Crystallinity from WAXS. Samples crystallized at 1 C/min.(d) Va n der Waals radii.(e) C X bond length.(f) CH 3 group on and every 21 st carbon 150 Table 2 2. Lattice parameters from WAXD for orthorhombic PE19X (X = H, O, F) crystals. PackingCell a ( ) b ( ) c ( ) V ( 3 ) Density (g/cm 3 ) PE Orthorh 7.410 4.934 2.547 93.131 0.9983 PE19O Orthorh 7.643 4.937 2.547 96.098 0.9748 PE19F Orthorh 7.540 5.032 2.547 96.639 0.9802 Table 2 3. T 1 H estimations from two component fits for PE19X (X = H, F, Cl, Br). Valu es in ms. Sample Fast Slow PE 2 25 PE19F 1.8 24 PE19Cl 3.8 6 PE19Br 2.2 8
72 Table 2 4. 13 C NMR chemical shifts for crystalline carbons of PE19X precision ADMET polyolefins (X = F, Cl, Br) and concentration of X groups in the crystal. Chemical Shift (ppm) Concentration of X in Crystal (mol %) (a) Sample CH2 (PE like) CH2 CH2 CHX From CH2 intensity From CH2 intensity From CHX intensity PE 32.8 PE19F 33.1 38.1 28.6 97 (b) 5.0 5.8 4.5 PE19Cl 34.1 41.0 ~31 67 5.2 5.8 5.5 PE19Br 34.1 41.5 N O (c) 62.5 5.0 4.7 (a) Calculated from ratio of integrated intensities in the crystalline spectra ( Figure 2 8 ). (b) Broad resonance. (c ) Not Observed Figure 2 1. Monomer and Polymer Synthesis.
73 Figure 2 2. TGA for P E19F (a) and PE19Br (b) showing the loss of HF and HBr respectively, followed by catastrophic decomposition.
74 Figure 2 3. IR spectra for thin films of PE19F PE19Cl and PE19Br cast on KBr disks.
75 Figure 2 4. WAXD diffractograms of linear PE and ADMET precisely substituted polyethylenes slowly cooled from the melt at ~1/min. The pendant group (X = H, O, F, Cl, Br) is indicated in each pattern.
76 Figure 2 5 DSC exotherms (a) and endotherms (b) of line ar PE and precisely substituted polyethylenes cooled from the melt at 10 C/min and further heated at 10 C/min. Figure 2 6 Peak melting temperatures of precisely substituted polyethylenes (PE19X) versus van der Waals radius of substituent (X). The linear regression was applied to halogenated samples.
77 Figure 2 7 Fully relaxed 19 F (a) and 13 C NMR (b, c) DD/MAS spectra of PE19F PE19Cl and PE19Br respectively. The Fit of the 19 F NMR spectral intensity with two Lorentzian components, corresponding to crystalline and non crystalline F environments is shown in the inset. Also shown are expansions of the 50 to 80 ppm regions of the 13 C NMR spectra fitted with two components for better distinction of CHCl and CHBr crystalline and non crystalline resona nces. [ (CH 2 ) 16 CH 2 ( ) CH 2 ( ) CHX ] y
78 Figure 2-8. Solid state 13C NMR. (a) Unfiltered 13C NMR CP MAS spectra and (b) 1H spin locking filtered (crystalline) spectra of pr ecision PE19X (X = F, Cl, Br) polyolefins.
79 CHAPTER 3 PRECISION ETHYLENE / VINYL CHLORIDE POLYM ERS VIA CONDENSATION POLYMERIZATION 3.1 Introduction As the second largest volume plastic produced in industry, 162 poly(vinyl chloride) (PVC) continues to find a growing range of applications and is project ed to have an increasing demand in the near future. 163 Common applications of PVC include pipes, siding, molding, windows, and flooring. 30 For many applications, the properties of PVC must be modified in order to increase performance and stability. While the addition of additives is the most common way to modify the characteristics of PVC for a specific application, the physical properties and stability are greatly affected by the primary structure of the PVC its elf. 22 For example, the amount of syndiotactic segments in PVC has a strong effect on the melting point and overall crystallinity of the polymer and the content of defects, such as branching, head to head linka ges, and mechanism responsible for its degradation at elevated temperatures. The occurrence of defects and the lack of control over primary structure is the consequenc e of the free radical initiated methods, which are used in the synthesis of nearly all PVC produced. 24 Based on this and the ubiquitous use of additives to modify properties in PVC, detailed information about t he effect of primary structure on materials properties is often not pursued. However, vinyl chloride copolymers are used as a means to access a variety of new property sets and to combine the attractive properties of PVC with other polymers, while minimizi ng the weaknesses of PVC. 164 Ethylene vinyl chloride (EVC) copolymers are an attractive class of polymers, which are pursued as a means of capitalizing on the strengths of PVC and polyethylene (PE). Here, judicious compositional control is a route to develo ping a series of polymers with accurately controlled properties as determined by the composition of the polymer itself. However, based on
80 the methods by which these EVC polymers are usually synthesized, little primary structure control is achieved and the more or less random structures give little information about how chlorine content and sequence distribution directly affect the materials properties of the polymers. There are three common methods for the preparation of EVC polymers: copolymerization of et hylene and vinyl chloride using free radical 165 166 167 or Ziegler Natta techniques, 75 77 reductive dechlorination of PVC, 104 105 106 and chlorination of polyethylene. 78 79 80 81 In the direct free radical copolymerization of vinyl chloride and ethylene, it is difficult to control the content of the two monomer units over the full range of compositions. 74 Further, the free radical nature of the polymerization leads to poorly defined structures with high defect contents. Direct copolymerization of vinyl chloride and ethylene using Ziegler Natta or metallocene catal ysts is also problematic based on the occurrence of side reactions with the activated chlorine atom on the monomer and the alkylaluminum cocatalyst as well as the tendency of the vinyl chloride monomer to undergo halo elimination after insertion into the metal alkyl bond. 75 77 In the case of the reductive dechlorination of PVC, while the ability to generate a family of EVC polymers with varying Cl contents and with identical chain lengths from a single PVC sample is a strong point for a model, 108 this class still suffers from defects present in the parent PVC. 107 The most common method use d for synthesizing EVC polymers is the chlorination of PE either under homogenous or heterogeneous conditions. In homogeneous, or solution based techniques, chlorine is distributed randomly throughout the polymer. 81 At low chlorine contents, the probability of chlorine substitution on a CHCl unit or neighboring unit is very small based on the bulky Cl atom, resulting in a random spatial distribution of Cl atoms along the backbone. At higher chlorine contents how ever, geminal chlorination and vicinal chlorination becomes more
81 commonplace. 89 90 91 In heterogeneous methods such as suspension, or solid state, the main drawback here is the blocky nature of the formed polymers based on the tendency for chlorination to occur in the amorphous regions of the polymer and at the edge of the crystalline domains. 80 93 Considering the above methods for attaining EVC polymers, it is clear that control over chorine content can be achieved in each of these methods, but control over sequence distribution cannot be attained Therefore, even though the properties of EVC polymers with different chlorine contents are known to vary, deriving exact relationships between primary structure and physical properties is unattainable because a precise distribution of comonomer units bet ween crystalline and amorphous domains has yet to be determined. 110 Despite the lack of a precise model for understanding the variation in the properties of EVC polymers, there is great practical interest in th ese materials. 110 Specifically, the introduction of ethylene units into a PVC chain will serve the role of an internal plasticizer, which can give many advantages over the introduction of an external plasticize r and lead to a new range of materials properties from a single component material. 74 111 From a more fundamental standpoint, this class of polymers is also of great interest for d eveloping a precise understanding of the relationship between molecular structure and physical properties in semicrystalline polymers in general, where EVC polymers could serve as an all purpose model for this class of polymers. 110 As described above however, no specific catalyst system has been developed to effectively copolymerize ethylene and vinyl chloride, leading to the necessity to look toward model EVC polymers to establish the fundamental significance o f this class of polymer. Such a focus on model polymers will not only help to derive structure property relationships, but will shift focus to the development of effective catalysts for the synthesis of EVC copolymers. 109 It is difficult however to find suitable model systems that will allow the development of precise
82 structure property relationships 112 and will not be limited by the quality of the synthetic or prepa rative method. Recently, olefin metathesis chemistry combined with post polymerization hydrogenation has emerged as an attractive route for synthesizing polyolefins bearing a variety of substituents. 113 For EVC polymers, ring opening metathesis polymerization (ROMP) has been employed to make a series of polymers with varying chlorine contents. 121 The absence of defects in polymers produced by ROMP leads to material s with well defined primary structures and chlorine contents that are tuned by controlling the ratio between chlorinated and non chlorinated monomers. The sequence distribution is nonetheless still random. Alternatively, acyclic diene metathesis polymeriza tion (ADMET) has been used to achieve an EVC polymer with both precisely controlled chlorine content and sequence distribution, 168 here such a level of control allows one to derive precise relationships between polymer primary structure and physical properti es in EVC polymers. Further, ADMET allows the investigation of a much greater range of polymers with precisely distributed chlorine atoms, as the linear dienes used in ADMET are free of the structural constraints required for monomers used in ROMP. We present here the full characterization of a new expanded family of precise EVC polymers in which a range of precise chlorine contents and sequence distri butions are examined in order to generate structure property relationships useful for developing a deeper understanding of the importance of primary structure control in EVC polymers. 3.2 Results and Discussion 3.2.1 Monomer and Polymer Synthesis Synthes is of the necessary chlorinated diene monomers is illustrated in Figure 3 1. The alcohol precursors ( 4 6 ) were prepared as described earlier. 169 Conversion of the alcohols to
83 the corresponding chlorine monomers ( 7 9 ) by reaction with carbon tetrachloride and triphenylphosphine preceded catalyst to yield the unsaturated ADMET polymer UPEXCl. The nomenclature acronym used herein is UPEXCl, where U indicates unsaturation and PEXCl indicates a polyethylene backbone with a chlorine substituent on every X th carbon, where X = 9, 15, 21 Exhaustive hydrogenation by diimide reduction 144 gave clean conversion to the fully saturated polymers PE9Cl PE15Cl and PE21Cl Polymer molecular weights are shown in Table 3 1 and vary between 31,000 and 51,000 g/mol. 3.2.2 Primary Structure Characterization The primary structure of these well defined UPEXCl and PEXCl polymers was established using a combination of 1 H NMR, 13 C NMR, TGA (thermogravimetric analysis), IR spectroscopy, and elemental analysis. Figure 3 2 shows the 1 H NMR and 13 C NMR spectra for a typical conversion of a symmetrical chlorine containing alpha omega diene monomer 7 6 chloroundeca 1,10 diene, to its unsaturated ADMET polymer UPE9C l and then to its precisely substituted saturated polymer analogue with a chlorine atom on each and every 9 th carbon PE9Cl Figure 3 2a and d illustrate the 1 H NMR and 13 C NMR spectra for the monomer 7 showing the assigned protons and carbons 1 through 6 in (5.0, 5.8, 2.1, 1.5, 1.7, 3.9 ppm) 1 H NMR and in (115.1, 138.6, 33.4, 25.9, 38.1, 64.0 ppm) the 13 C NMR respectively. Figure 3 2b and e illustrate the 1 H NMR and 13 C NMR spectra for the unsaturated ADMET polymer UPE9Cl In the proton spectrum (Figure 3 2b) two triplets are observed in the region 5.34 5.40, corresponding to the cis and trans configurations of the double bond. The 13 C satellite of the major peak is a doublet of triplets, with a coupling constant of 17.5 Hz, therefore the major peak was as signed to trans The trans : cis ratio determined from the carbon spectrum (Figure 3 2e), is 4 : 1. The peaks for the
84 terminal vinyl moiety appeared at 4.99 ( CH 2 trans ), 4.93 ( CH 2 cis ) and 5.81 ( CH ). The degree of polymerization (DP) was determined by inte gration of these signals against the two triplets in the region 5.34 5.40 and the values are shown in Table 3 2. The proton signals for UPE9Cl were assigned from the DQCOSY spectrum. The protons in positions 1 and 2 displayed different chemical shifts in t he trans and cis moieties, being 0.03 ppm upfield and 0.03 ppm downfield in cis as compared to trans in positions 1 and 2, correspondingly. The GHMBC spectrum revealed both the one bond and the two or three bond couplings between protons and carbons, for both the cis and the trans moieties. The proton and carbon chemical shifts have been assigned based on these cross peaks and are presented in Table 3 2, together with the trans : cis ratio. The differences between the chemical shifts of a carbon atom in the trans and cis moieties, trans cis = trans cis have been measured in a 13 C spectrum acquired with a digital resolution of 0.9 ppb and are given, in ppb in Table 3 2. trans cis are noticeable for carbons up to four bonds away from the double bond, and for an alkyl moiety f ree of the influence of other groups (double bonds or chlorine atoms must be some 7 10 bonds away) typical values in the order of distance, starting with the carbon of the double bond, are: 465, 5339, 110, 137, 37 ppb. Further, in Figure 3 2c the 1 H NMR for PE9Cl shows a methine proton ( at ppm, and methylene protons ( =4) at 1.7 ppm and (1 3) at 1.2 1.6 ppm. The 13 C NMR data exhibited in Figure 3 2f reveal that five sp 3 carbon signals are present in this polymer, which contains a methine carbon ( ppm and methylene carb ons ( =4, 38.8 ppm), ( =2, 29.6 ppm), ( =1, 29.4), ( =3, 26.7). The chemical shift range observed for polymer PE9Cl are in very good agreement with the values in the literature for the random ethylene/vinyl chloride copolymers. 104 121 125 170 In the random case, several different methine and other methylene chemical shifts are observed due to the random nature of the polymer structure. Obtaining only fi ve different
85 carbons for ~49,000 molecular weight polymers unequivocally confirms the precise structure of these polymers. These spectral data not only support the primary structure of the repeat unit but also suggest that no side reactions are detectable within the limitations of the NMR instrument. Elemental analysis results also show good agreement between theoretical and experimental values (see experimental section). The olefin region in these NMR spectra also illustrates the conversion of monomer to unsaturated polymer with the disappearance of the terminal olefin at 5.0 and 5.8 ppm in the monomer and the subsequent growth of internal olefin resonance at 5.4 ppm in the unsaturated polymer UPE9Cl a result consistent with the GPC results given in Table 3 1. Upon exhaustive hydrogenation, these olefin resonances completely disappear. The 13 C NMR spectra support exhaustive hydrogenation as well; note that the sp 2 resonances in the unsaturated polymer ( trans 130.2 ppm, cis 129.7 ppm) completely disappear after hydrogenation giving PE9Cl (confirmed by IR). These spectra are typical for all the ADMET polyethylenes synthesized in the series, and they illustrate the degree of structure control that is possible. These conclusions are valid for the UPE15Cl UPE 21Cl polymers as well; the chemical shift values are given in Table 3 2 and in the experimental section. This detailed NMR study confirms the precise nature of the examined polymers and the degree of absolute control over primary structure afforded by ADME T polymerization followed by exhaustive hydrogenation. The IR spectra in Figure 3 3 display characteristic information regarding the primary structure of the polymers, where the absence of a peak at 967 cm 1 for all three EVC polymers indicates complete hy drogenation, based on the disappearance of the out of plane olefin C H wagging vibrational mode. The first spectral segment with characteristic information is the C C1 stretching region between 600 and 700 cm 1 For PVC, three broad bands are observed cent ered
86 at 615, 636, and 693 cm 1 148 of different conformations and configurations. For the EVC polymers, the resonances in the 600 700 cm 1 regi on are significantly narrowed relative to PVC as the number of chlorines is reduced and number of CH 2 sequences increased. The spectra then consist of two single narrow bands at 660 661 and 609 611 cm 1 which have been assigned to isolated gauche, and tra ns C C1 stretching respectively. 126 Going from PE9Cl to the PE21Cl decreases the gauche to trans ratio as the number of CH 2 units between Cl atoms increases. This observation is significant because it shows a distortion of the all trans trans zigzag conformation of the backbone at higher chlorine contents. Another region of particular interest is the 700 850 cm 1 region where methylene rocking motions are observed. The resonance at 758 cm 1 has been assigned to the methylene rocking motion in PVC, specifically the rocking mode of a CH 2 unit between two CHCl units. As the concentration of CH 2 units increases going from PVC through the EVC series, the peak at 758 cm 1 decreases in intensity and the peak at 723 718 cm 1 increases in intensity, which is characteristic of CH 2 rocking modes for longer methylene sequences. In the case of PE a doublet is observed at 730 and 719 cm 1 which has been assigned to rocking vibrations of long methylene sequences. This doublet in PE, which is specifically associated with the orthorhombic crystal structure, is not observed for any of the EVC polymers suggesting that a different crystalline structure exist for these materials. 148 Band s between 1400 and 900 cm 1 are difficult to assign to individual vibrational components, since most resonances overlap with other resonances. The numerous resonances in this region are rapidly reduced in intensity as the number of chlorines is decreased. Exceptions are the 1332 and 963 cm 1 bands which are dominated by C C stretching and CH 2 wagging motions of short sequences. In the case of EVC polymers, the peak at 1332 cm 1 decreases with
87 decreasing chlorine content and the peak at 963 cm 1 fully disapp ears for all of the EVC polymers. Additionally, all the peaks in the area of the 1254 cm 1 multiplet arise from methine bending modes and are sensitive to sequence length and/or environment. The decrease in the intensity of the 1254 cm 1 peak with decreasi ng chlorine content is indicative of the smaller chlorine contents and the changing environment caused by the presence of longer methylene sequences. 148 The C H bending modes at 1435/1426 cm 1 for PVC broaden into a singlet as the number of chlorines is decreased. Concomitant with the 1435/1426 cm 1 loss is the growth of the 1467 1471 cm 1 resonance which eventually becomes the 1472/1462 cm 1 doublet assigned to the methylene bending modes in PE. 148 The shoulders observed in the EVC samples at 1457 and 1434 cm 1 correspond to the bending modes for the and methylenes. It can be observed that the intensity of the peaks at 1457 and 1434 cm 1 increases with increasing chlorine content in the EVC polymers. The above IR study clearly illustrates the evolution of the PVC structure through a series of prec ise EVC polymers with decreasing chlorine content to the pure PE backbone. 3.2.3 Thermal Analysis TGA While NMR and IR support the proposed primary structure, TGA results directly support the precise composition of the polymers. Figure 3 4a displays the thermal decomposition curves of all three polymers as shown relative to PE and PVC. All three EVC polymers exhibit a two stage decomposition, where the first stage corresponds to the loss of HCl and the second stage marks the catastrophic decomposition of the polymer. Analogous to our previous work on ethylene vinyl halide polymers and the reported decomposition of PVC, the mass loss in the first stage quantitatively reflects the halogen content of the polymer. The observed values for mass
88 loss in the first stage are found to be in agreement with the calculated HCl content for each of the EVC polymers at 23%, 15%, and 11% for PE9Cl PE15Cl and PE21Cl respectively. Figure 3 4a also shows that the onset of decomposition for the first stage loss of HCl increas es with decreasing chlorine content (see Table 3 1). The EVC samples become more stable with increasing content of ethylene units as the labile chlorine content decreases, since the chain reaction (zipper mechanism) resulting in HCl loss is restricted by m ore ethylene units compared to PVC. This effect is called inner stabilization. 105 The stabilization of the EVC samples can also be partially attributed to the lack of tertiary or allylic chlorines, which are ge nerally found in PVC. However in the second stage of decomposition, the onset for PVC is higher than that for the EVC polymers and this difference can be attributed to a high concentration of conjugated carbon carbon double bonds that begin to crosslink an d are converted into char. 171 Even at temperatures above 500 C, after the EVC polymers have completely decomposed, residual char is observed in PVC and a non zero mass is observed in PE, which is attributed to catalyst residue. Figure 3 4b compares the sta bility of the EVC polymers to that of PVC under isothermal decomposition at 180 C. Here, a temperature of 180 C is chosen as it is below the onset of decomposition for any of the polymers, ensuring that any mass loss may be attributed to HCl loss. The p ercentage HCl loss is normalized to the overall HCl content in each of the polymers samples undergo a rapid loss of 3.5 6.8% of the total HCl content of the polyme r during the first hour. However, after the first hour, the degradation rate of the EVC polymers levels off leading to a total HCl loss of only ~6% even after four hours at 180 C, while the PVC sample continues to degrade at nearly a constant rate, leadin g to the loss of ~40% of the HCl content after four
89 hours. Structural precision and the absence of defects make the difference. Precise EVC polymers are more stable to the applied isothermal decomposition than PVC, a result which can be attributed to the s uppression of the zipper mechanism found in PVC and induced by the extended spacing between chlorine atoms (referred to as inner stabilization as described above) in the precise polymers. 3.2.4 DSC The DSC data for the precision polymers is shown relative to that for PE and PVC in Figure 3 5 for both the crystallization (Figure 3 5a) and melting (Figure 3 5b) behavior, and the results are summarized in Table 3 1. Increasing the chlorine atom content results in a decrease of T m and H m This behavior indicates an increase in the degree of disorder in the crystal owing to the introduction of the chlorine atom into the crystal lattice. When we compare these results with our previous work on precision ethylene/vinyl bromide polymers, 169 it becomes clear that Br atoms lower the melting point and H m more than the chlorine atoms for polymers with the same halogen content. Assuming a similar distribution of the halogen (Cl or Br) between crystal line and non crystalline regions, 168 the difference in thermal behavior correlates with larger lattice strains in the precision vinyl bromide series. The sharp crystalline and melting traces seen in Figure 3 5 are typical of homopolymers and are in stark contrast to the broad traces observed in random EVC polymers of similar chlorine content. 74 This suggests a difference in the crystallization process for ADMET EV C polymers relative to random EVC polymers, which is not based on the selection of long crystallizable sequences. It is likely that the crystallization of precisely substituted Cl samples evolves by incorporating in the crystallite long chain segments with out chlorine discrimination. 168 PE9Cl displays a sharp crystallization peak and a dual melting transition, the
90 latter caused by a melting recrystallization process enabled by the presence of relatively high Cl contents in the crystal ( Figure 3 6 ). The lower melting peak corresponds to the initial crystallites, while crystals formed during the heating scan melt at a higher temperature. Recrystallization is suppressed at the heating rate of 40 o C/min Reducing th e Cl content in the series, and hence in the crystal, leads to less defected, thicker crystallites not subject to melt recrystallization as indicated in Figure 3 5 by sharp single meltings of PE15Cl and PE21Cl. The dramatic decrease of heat of fusion for P E9Cl relative to PE15Cl and PE21Cl suggests a lower degree of crystallinity, indicating that the increased content of bulky atoms along the chain inhibits the ability of the polymer to crystallize to a greater extent. The greater amorphous content of PE9Cl is evidenced by a more visible T g relative to the other precise polymers. Interestingly we observe a linear correlation between the T g of PVC at 80 C, the T g of PE at 120 C, 172 and the value for PE9Cl at 33 C with mass fraction of Cl in the polymer. The glass transition temperature of PE9Cl decreases with decreasing chlorine content relative to PVC due to the high mobility of the ethylene units. Figure 3 7 shows the linear relationship between T m and the number of CH 2 groups in the repeating unit. In accord with the behavior of trans poly(alkenamers), the melting temperature increases with increasing number of carbon atoms in the repeating unit for materials that crystallize with the same crystal structure, as is expected in this case. 3.3 Conclusion a nd Outlook Herein we have described the synthesis of a family of precise ethylene/vinyl chloride polymers that contain a chlorine atom on each and every 9 th 15 th and 21 st carbon respectively. The precise primary structure has been confirmed by 1 H NMR and 13 C NMR spectroscopy, which allow for the assignment of each chemical shift observed to one unique methylene or methine group within the repeat unit. The presence of only those shifts corresponding to a
91 precision polymer reinforces the ability of ADMET to generate precise and defect free polymers. The thermal characterization of the polymers via DSC suggests a crystallization process typical of homopolymers as evidenced by the sharp melting and crystallization transitions. The homopolymer behavior is in dicative of a precise structure that crystallizes in such a fashion to incorporate the chlorine atoms into the crystal lattice, in sharp contrast to random EVC polymers. Melting and crystallization temperatures scale proportionally to the number of CH 2 gro ups in the repeating unit. The heat of fusion decreases drastically with increasing Cl content in the series, with values of < 30 J/g for PE9Cl suggesting that beyond a certain critical chlorine content in these precise polymers, the bulky chlorine atoms b egin to inhibit the crystallization process. Further work aimed to provide direct evidence of the crystal structure via WAXD, crystal composition via solid state NMR and lamellar morphology is currently underway. 3.4 Experimental Chemicals. Chemicals were purchased from the Aldrich Chemical Company and used as benzylidine ruthenium (IV)dichloride, was purchased from Strem Chemical and stored in an Argon filled dry box prior to use. Methylene chloride and o xylene were distilled over CaH 2 The same ADMET PE sample published 145 by our group was used for comparison. PVC was purchased from the Aldrich Chemical Company, P roduct Number 389293 Instrumentation. Solution 1 H NMR (300 MHz) and 13 C NMR (75 MHz) spectra were recorded on a Varian Gemini 300, VXR 300 or Mercury 300 spectrometer. All chemical shifts for 1 H and 13 C NMR spectroscopy were referenced to residual signals from CDCl 3 ( 1 H = 7.27 ppm and 13 C = 77.23 ppm) with an internal reference TMS 0.03% v/v to internal TMS standard for 0. The UPE XCl polymers were also characterized by standard 2D NMR experiments run on
92 a Varian Inova 500 instrument (500 MHz for 1 H and 125 MHz for 13 C). In all the NMR work the solvent was chloroform d and the temperature was 25 C. H igh resolution mass spectral (HRMS) data were obtained on a Finnegan 4500 gas chromatograph/mass spectrometer using the electron ionization (EI) mode. Element al analyses were carried out by Atlantic Microlabs Inc., Norcross, GA. The GPC measurements for samples in THF were taken on a Waters GPCV 2K instrument Samples were run with HPLC grade THF at 40 C on Waters Styrage1HR 5E columns monitored with an intern al differential refractive index detector (DRI) relative to polystyrene standards Polymer molecular weights reported vs polyethylene standards were measured using a Waters Associates 150C high temperature gel permeation chromatograph equipped with three P olymer Laboratories mixed bed Type B columns and an internal DRI detector. The mobile phase was BHT inhibited 1,2,4 trichlorobenzene (135 C, flow rate 1.0 mL/minute, typical sample concentration 2 mg/mL). IR data was obtained using a Perkin Elmer Spectrum One FTIR outfitted with a LiTaO 3 detector. Measurements were automatically corrected for water and carbon dioxide. Thermogravimetric analysis (TGA) data was obtained with a Perkin Elmer 7 series thermal analysis system. The TGA samples (2 5 mg) were heate d from 50 C to 700 C at 10 C/min under nitrogen. Melting and crystallizations were obtained at 10 C/min in a differential scanning calorimeter TA Instrument DSC Q1000 V9.6 Build 290 under nitrogen flow and calibrated with indium. Synthesis: General pro cedure for Grignard reaction. Synthesis of 5 bromopent 1 ene ( 1 ), 8 bromooct 1 ene ( 2 ), 11 bromoundec 1 ene ( 3 ) undeca 1,10 dien 6 ol ( 4 ), heptadeca 1,16 dien 9 ol ( 5 ), and tricosa 1,22 dien 12 ol ( 6 ) was described previously. 169 General procedure for chlorination reaction. The precursor alcohol 4 5 or 6 (1 equiv.) and Ph 3 P (1.5 equiv.) were dissolved in CCl 4 under argon. The reaction was stirred overnight at
93 80 C. The solution was concentrated, and pentane w as added. The mixture was filtered, and the filtrate was concentrated under vacuum to yield chlorinated ADMET monomers 6 chloroundeca 1,10 diene ( 7 ), 9 chloroheptadeca 1,16 diene ( 8 ), and 12 chlorotricosa 1,22 diene ( 9 ). The chlorinated ADMET monomers were purified with column chromatography on silica gel eluted by hexane (yield 70 80%). 6 chloroundeca 1,10 diene (7). 1 H NMR (300 MHz CDCl 3 ) 5.81 (m, 2H), 5.00 (m, 4H), 3.91 (p, 1H), 2.09 (m, 4H), 1.81 1.42 (br, 8H). 13 C NMR (75 MHz, CDCl 3 ) 138.55, 115.09, 64.00, 38.08, 33.40, 25.90. HRMS calcd. for C 11 H 19 Cl (M+Cl) 221.0858; found, 221.0866. Anal. Calcd. for C 11 H 19 Cl: C, 70.76; H, 10.26; Cl, 18.99. Found: C, 70.93; H, 10.25; Cl, 18.74. 9 chloroheptadeca 1,16 diene (8). 1 H NMR (300 MHz CDCl 3 ) 5.81 (m, 2H), 4.97 (m, 4H), 3.90 (p, 1H), 2.05 (m, 4H), 1.70 (m, 4H), 1.60 1.20 (br, 16H). 13 C NMR (75 MHz, CDCl 3 ) 139.31, 114.44, 64.52, 38.71, 33.96, 29.24, 29.19, 29.04, 26.66. HRMS calcd. for C 17 H 31 Cl (M + ), 270.2114; found, 270.2112. Anal. Calcd. for C 17 H 31 Cl: C, 75.38; H, 11.54; Cl, 13.09. Found: C, 75.53; H, 11.52; Cl, 13.00. 12 chlorotricosa 1,22 diene (9). 1 H NMR (300 MHz CDCl 3 ) 5.81 (m, 2H), 4.97 (m, 4H), 3.89 (p, 1H), 2.04 (m, 4H), 1.71 (m, 4H), 1.60 1.20 (br, 28H). 13 C NMR (75 MHz, CDCl 3 ) 139.45, 114.33, 64.61, 38.75, 34.03, 29.73, 29.72, 29.68, 29.41, 29.35, 29.16, 26.72. HRMS calcd. for C 23 H 43 Cl (M + ), 354.3053; found, 354.3066. Ana l. Calcd. for C 23 H 43 Cl: C, 77.81; H, 12.21; Cl, 9.99. Found: C, 77.88; H, 12.22; Cl, 10.00. General procedure for bulk polymerization. catalyst were combined in a ratio of 500:1 under argon atmosphere. The polymerizati on was conducted at 35 40 C under vacuum with stirring for 5 days. The reaction was then stopped and 5 mL of toluene was added to dissolve the polymer with stirring. The reaction was allowed to
94 cool to room temperature. The polymers were then precipitated by dripping the toluene solution into cold acidic methanol. They were then isolated by filtration and dried. Polymers were then redissolved in 50 mL of toluene and treated with THP (tris(hydroxymethyl)phosphine)) in order to remove any residual catalyst. 173 The polymers were then reprecipitated into acidic methanol, filtered and dried. Polymerization of 6 chloroundeca 1,10 diene (UPE9Cl). M w (GPC vs. PS) = 37,100 g/mol. PDI = ( M w / M n ) = 1.84. For 1 H NMR and 13 C NMR data see Table 3 2. Polymerization of 9 ch loroheptadeca 1,16 diene (UPE15Cl). M w (GPC vs. PS) = 39,800 g/mol. PDI = ( M w / M n ) = 1.81. For 1 H NMR and 13 C NMR data see Table 3 2. Polymerization of 12 chlorotricosa 1,22 diene (UPE21Cl). M w (GPC vs. PS) = 67,100 g/mol. PDI = ( M w / M n ) = 1.84. For 1 H NMR and 13 C NMR data see Table 3 2. General Procedure for hydrogenation. The chlorine containing polymers ( UPE9Cl ), ( UPE15Cl ), and ( UPE21Cl ) were then hydrogenated using a modified version of the method described by Hahn 144 by dissolving in dry o xylene under argon and adding 3.3 equivalents of p toluenesulfonyl hydrazide (TSH) and 4 equivalents of tri n propyl amine (TPA). The solutions were refluxed for 9 hours and then cooled to room temperature. The hydrogenate d polymer was precipitated into ice cold methanol and isolated by filtration. The dried polymer was then redissolved in toluene and re precipitated by dipping into ice cold acidic methanol. A white solid was collected by filtration and the polymers were is olated in quantitative yield. PE9Cl. 1 H NMR (300 MHz CDCl 3 ) 3.89 (p, 1H), 1.71 (m, 4H), 1.62 1.0 (bm, 12H). 13 C NMR (75 MHz, CDCl 3 ) 64.58, 38.75, 29.64, 29.37, 26.71. Anal. Calcd.: C, 67.27; H, 10.66; Cl, 22.06. Found: C, 67.25; H, 10.65; Cl, 20.92. M w (GPC vs. PS) = 48,700 g/mol. PDI = ( M w / M n ) = 1.78.
95 PE15Cl. 1 H NMR (300 MHz CDCl 3 ) 3.90 (p, 1H), 1.71 (m, 4H), 1.6 1.2 (bm, 24H). 13 C NMR (75 MHz, CDCl 3 ) 64.63, 38.76, 29.90, 29.87, 29.82, 29.76, 29.43, 26.74. Anal. Calcd.: C, 73.58; H, 11.94; Cl, 14.48. Found: C, 73.66; H, 12.00; Cl, 14.26. M w (GPC vs. PS) = 51,400 g/mol. PDI = ( M w / M n ) = 1.75. PE21Cl. 1 H NMR (300 MHz CDCl 3 ) 3.90 (p, 1H), 1.74 (m, 4H), 1.6 1.2 (bm, 36H), 0.9 (t, 0.15H). 13 C NMR (75 MHz, CDCl 3 ) 64.62, 38.76, 29.94, 29.88, 29.82 29.76, 29.44, 26.74. Anal. Calcd.: C, 76.66; H, 12.56; Cl, 10.78. Found: C, 76.64; H, 12.61; Cl, 10.57. M w (GPC vs. PE) = 31,100 g/mol. PDI = ( M w / M n ) = 4.43.
96 Table 3 1. Molecular weight and thermal data for unsaturated (UPEXCl) and satura ted (PEXCl) polymers. Sample M w x10 3 M w \ M n T g (C) T m (C) T c (C) H m (J/g) First stage onset of decomposition e (C) Second stage onset of decomposition f (C) UPE9Cl 37 a 1.84 63 292 385 UPE15Cl 40 a 1.81 57 24 6 27 320 413 UPE21Cl 67 a 1.84 37 46 32 67 332 415 PVC 43 1.95 80 274 450 PE9Cl 49 a 1. 78 33 41 15 26.7 295 418 PE15Cl 51 a 1.75 63 54 87.4 301 420 PE21Cl 31 b 4.43 81 70 111.3 327 407 PE 16 c 1.60 120 d 129 118 185.0 348 384 a.) GPC vs. PS in THF; b.) GPC vs. PE in DCB; c.) ref 145 ; d. ) ref 174 ; e.) Recorded at first stage 10% total mass loss under nitrogen gas, 10 C/min.; f.) Recorded at second stage 10% total mass loss under nitrogen gas, 10 C/min. Table 3 2. Proton and carbon chemical shifts (in ppm) and trans cis (in ppb) for polymers UPE9Cl UPE15Cl and UPE21Cl comp ound trans:cis position 1 2 3 4 5 6 7 8 9 10 11 DP UPE9Cl 5.0 C 130.20 32.00 26.40 37.90 64.00 trans cis 514 5334 148 86 31 100 H 5.40 1.99, 1 .99 1.59, 1.46 1.70, 1.70 3.89 UPE15Cl 4.4 C 130.30 32.50 29.50 29.00 29.10 26.50 38.50 64.30 trans cis 475 5387 115 136 41 0 0 14 72 H 5.38 1.97, 1.97 1.34, 1.34 1.27, 1.27 1.33, 1.30 1.51, 1.39 1.69, 1.69 3.88 UPE21C l 4.1 C 130.30 32.60 29.70 29.20 29.49 29.56 29.52 29.20 26.50 38.50 64.30 trans cis 465 5399 110 137 37 0 0 0 0 0 0 148 H 5.38 1.96, 1.96 1.33, 1.33 1.27, 1.27 1.27, 1.27 1.27, 1.27 1.27, 1.27 1.27, 1.27 1.51, 1.39 1.69, 1.69 3.88
97 Figure 3 1. Synthesis of PEXCl polymers.
98 Figure 3 2. 1 H NMR spectra for monomer 7 (a), UPE9Cl (b), PE9Cl (c) and 13 C NMR spectra for monomer 7 (a), UPE9Cl (b), PE9Cl (c).
99 500 750 1000 1250 1500 1750 PVCWavenumber ( cm-1 ) Transmittance (a. u.) 967 cm-1No C=C-H PE9Cl PE15Cl PE21Cl PE Figure 3-3. IR of PE, PE21Cl, PE15Cl, PE9Cl, and PVC.
100 Figure 3 4. TGA results for PE, PE21Cl PE15Cl PE9Cl and PVC. (a) TGA results for PE, PE21Cl PE15Cl PE9Cl and PVC. (b) Normalized HCl loss data for PVC, PE9Cl PE15Cl PE21Cl under isothermal (180 C) degradation measured under nitrogen atmosphere.
101 Figure 3 5. DCS exotherms (a) and endotherms (b) of PE9Cl PE15Cl PE21Cl relative to PE and PVC. Figure 3 6 DSC endotherms for PE9Cl at heating rates of 5, 10, and 40 C/min.
102 Figure 3 7 Trends for the variation in T m vs. the number of CH 2 per repeat unit.
103 CHAPTER 4 PRECISION ETHYLENE/V INYL BROMIDE POLYMER S 4.1 Introduction The importance of polyolefins in industry is reflected by the enormous volumes of these polymers that are produced on an annual basis. 116 Wh ile polyethylene (PE) is the most abundant of these polymers, polyethylenes modified with halogens such as poly(vinyl chloride), are extensively employed as a result of the tailored sets of properties they possess. 118 175 Ethylene/vinyl bromide (EV/B) polymers have received significantly less attention than their ethylene/vinyl fluoride (EVF) or ethylene/vinyl chloride (EVC) counterparts because they are difficult to synthesize, especially those having a well defi ned structure. 123 No direct synthetic copolymerization methods exist that combine vinyl bromide and comonomers; 107 instead a variety of postpolymerization brominat ion strategies have been employed. The most common techniques employ either bromination of a suspension of PE, or solution bromination. 102 Similarly, solution hydrobromination of polyoctenamer has been used. 176 Additionally, surface bromination of thin PE films with bromine vapor has been described. 177 Another somewhat different approach to these polymers involves reductive debromination of poly(vinyl bromide). 107 Defi ciencies in the above techniques can be attributed to the incorporation of structural defects in the polymer precursor (e.g. branching, cyclic defects caused by backbiting) as well as the uncontrolled chemical events that accompany postpolymerization bromi nation. 107 Further, the uncontrolled incorporation of bromine, which is assured when using these techniques, leads to polymers with a non uniform distribution of bromine or even a blocky primary structure. 102 Consequently, the pursuit of optimized routes to well defined EV/B copolymers remains relevant. For example, EV/B copolymers could serve as the precursor to a variety of modified
104 polyolefins via substitution of the bromine atom 178 179 or, they could prove useful in selective crosslinking. In addition, the EV/B copolymers could find direct materials applications stemming from the flame retardancy induced in bromine containing materials, 180 or more generally based on the modified physical properties and their influence on adhesion, wettability, and surface energy. 177 Olefin metathesis offers distinct advantages over previously used methods for the synthesis of EV/B polymers. Acyclic diene metathesis polymerization (ADMET) is now widely accepted as a route to polyolefins having a precisely defined primary structure, for it entails mild chemistry, which does not lead to side reactions involving the reactive bromine functionalit y. Thus far we have produced a precise EV/B ADMET polymer with a bromine atom located on each and every 19 th carbon along the polyethylene backbone. 168 We now extend this study to a larger family of EV/B polym ers in order to examine the effect of bromine content and sequence distribution on the thermal and crystalline properties of this novel class of polymers. 4.2 Results and Discussion Synthesis of the necessary brominated ADMET diene monomers is illustrated in Figure 4 1. The alcohol precursors ( 4 6 ) were prepared according to the methods previously described 181 and converted to the corresponding bromine monomers ( 7 9 ) by reaction with carbon tetrabromide and triphenylphosphine 168 Bulk generation ruthenium catalyst, followed by exhaustive hydrogenation under mild conditions with PE9Br PE15Br an d PE21Br containing a bromine atom on each and every 9 th 15 th and 21 st carbon along the PE backbone, respectively. The primary structure of the polymers was confirmed by 1 H NMR, 13 C NMR, and elemental analysis.
105 The primary structure of these polymers was confirmed further by thermogravimetric analysis (TGA) (Figure 4 2a) and infrared spectroscopy (Figure 4 2b). All three polymers exhibit a two stage TGA weight loss where the first stage corresponds to the loss of HBr and the second stage marks the catastr ophic decomposition of the polymer. Analogous to our previous work on ethylene vinyl halide polymers 168 and the reported decomposition of poly (vinyl bromide), 180 182 the mass loss in the first stage quantitatively reflects the halogen content of the polymer. The observed values for mass loss in the first stage are found to be in agreement with the calculated HBr content for each of the polymers at 39%, 28%, and 22% fo r PE9Br PE15Br and PE21Br respectively. The IR spectra also display characteristic information regarding the primary structure of the polymers. The absence of a peak at 967 cm 1 for all three polymers indicates complete hydrogenation, based on the disapp earance of the out of plane olefin C H wag, observed in the unhydrogenated polymers UPE9Br UPE15Br and UPE21Br (Figure 4 4). Characteristic peaks corresponding to C Br stretching are found in the region from 700 450 cm 1 and in all cases the peaks at ~61 3 cm 1 and ~538 cm 1 can be assigned to secondary bromines in the gauche and trans configuration respectively. 183 From these data it becomes clear that with increasing bromine content, the ratio of gauche C Br (~613 cm 1 ) increases relative to trans C Br (~5 38 cm 1 ). Corresponding methylene peaks at 1368 cm 1 and 1308 cm 1 also indicate the presence of gauche and trans conformations, specifically the symmetric and antisymmetric wagging modes of gauche trans gauche triads. 101 Other characteristic methylene peaks are assigned for all polymers at ~1470 cm 1 and ~1432 cm 1 for methylene bending modes, 184 185 and at ~720 cm 1 for methylene rocking. 186 Importantly these spectra suggest, in accord with our previous work on
106 PE19 Br, that these three precise E/VB polymers possess a triclinic rather than an orthorhombic crystal structure based on the absence of doublets at 1463 1472 cm 1 and 721 730 cm 1 168 184 The proposed triclinic structure is also supported by differential scanning calorimetry (DSC) as well (Figure 4 3). Melting points (T m ) observed on the second cycle of heating and cooling are plotted relative to the bromine content of the polymers (Figure 4 3a). A linear decrease in T m is evident with increasing bromine content. This linear trend has been observed for a series of precise methyl branched PE derivatives synthesized via ADMET 128 (Figure 4 5); and here, the correlation of the trends is accentuated by the similarity in the size of the bromine atom and the methyl group The decrease in T m can be explained using the same arguments proposed for PE19Br and similar halogenated ADMET p olyolefins, 168 specifically, that increased bromine content in the crystal is the direct consequence of polymer composition and the larger incorporation of defects leads to a proportionally decreasing T m Bromi ne side groups were also found to enter the crystals of random EV/B polymers. 103 Figure 4 3b shows the relationship between the enthalpy of fusion ( H f ) and the bromine content. Here the H f (on a weight basis) shows a dramatic decrease as bromine content is increased at low mole fractions, while increases in bromine content at high mole fraction show markedly smaller changes in H f The initial large decrease in H f for PE21Br relative to PE reflects a predicted change in the crystal structure from orthorhombic to triclinic as we previously observed with PE19Br. 168 Owing to the participation of multiple r epeating units in the crystalline unit cells of the brominated ADMET polyolefins, most useful will be the comparison of thermodynamic H f data, per mole of crystalline repeating unit, for the same triclinic crystal structures. Since these values are at th e present time unknown, we use the observed heats of melting to infer trends with length of repeating unit. From the observed values, 17989 J/mol ( PE21Br ), 10140 J/mol
107 ( PE15Br ) and 4248 J/mol ( PE9Br ), H f scales proportionally to the number of CH 2 groups in the repeating unit. The value of PE9Br is nearly one fourth of H f of PE21Br denoting that as the crystals of the same structure incorporate more defects (bromine), the energy required to disorder the all trans packing of the backbone repeating unit i s dramatically decreased. Concomitantly the associated entropy of melting (in J/mol K) decreases from 52.5 J/mol K ( PE21Br ) to 16.4 J/mol K ( PE9Br ) following the expected impact of Br side groups in reducing the configurational entropy of the melt (Tabl e 4 1). Comparatively, the crystalline properties of this series of precision brominated polyethylenes follow the behavior of trans poly(alkenamers) for which the melting temperatures increase with the number of carbon atoms in the repeating unit for the s ame crystal structure. The enthalpy and entropy of fusion increase with the size of the repeating unit. However, the value of S per single bond remains effectively constant. 187 188 It appears that this feature may also hold for the precision Br samples as the calculated average S per bond in the series is 2.2 0.3 J/ mol K. 4.3 Conclusion The results presented here illustrate that metathesis chemistry is an efficient method for the synthesis of precisely defined EV/B polymers. The regular structure of t hese materials is reflected in the primary structure characterization and thermal analysis indicating strong correlations between properties and bromine content. Further work will serve to definitively establish the crystal and phase structure of these pol ymers utilizing WAXS and solid state 13 C NMR techniques, and also to begin to define potential applications for such polymers.
108 4.4 Experimental Chemicals. Chemicals were purchased from the Aldrich Chemical Company and used as first generation ruthenium catalyst, bis(tricyclohexylphosphine) benzylidene ruthenium (IV)dichloride, was purchased from Strem Chemical and stored in an 3 ) 3 was purchased from Strem. Methy lene chloride and o xylene were distilled over CaH 2 The same ADMET PE sample published 145 by our group was used for comparison. Instrumentation. Solution 1 H NMR (300 MHz) and 13 C NMR (75 MHz) spectra were reco rded on a Varian Associates Gemini 300, VXR 300 or Mercury 300 spectrometer. All chemical shifts for 1 H and 13 C NMR spectra were referenced to residual signals from CDCl 3 ( 1 H = 7.27 ppm and 13 C = 77.23 ppm) with an internal reference TMS 0.03% v/v. H igh re solution mass spectral (HRMS) data were obtained on a Finnegan 4500 gas chromatograph/mass spectrometer using the electron ionization (EI) mode. Elemental analyses were carried out by Atlantic Microlabs Inc., Norcross, GA. The GPC measurements for samples in THF were taken on a Waters GPCV 2K instrument Samples were run with HPLC grade THF at 40 C on Waters Styrage1HR 5E columns relative to polystyrene standards IR data was obtained using a Perkin Elmer Spectrum One FTIR outfitted with a LiTaO 3 detector. Measurements were automatically corrected for water and carbon dioxide. Thermogravimetric analysis (TGA) data was obtained with a Perkin Elmer 7 series thermal analysis system. The TGA samples (2 5 mg) were heated from 50 C to 700 C at 10 C/min. Meltin g and crystallizations were obtained at 10 C/min in a differential scanning calorimeter TA Instrument DSC Q1000 V9.6 Build 290 under nitrogen flow and calibrated with indium.
109 Synthesis: General procedure for Grignard reaction. 5 bromopent 1 ene ( 1 ), 8 bro mooct 1 ene ( 2 ), 11 bromoundec 1 ene ( 3 ) were synthesized using a published procedure. 189 Bromine compounds 1 2 3 were then used as a reagents for Grignard reaction in the presence of Mg and formaldehyde in THF using the same methodology as reported in the literature 181 to yield precursor alcohol compounds undeca 1,10 dien 6 ol ( 4 ), heptadeca 1,16 dien 9 ol ( 5 ), and tricosa 1,22 dien 12 ol ( 6 ). The alcohol precursors were purified with column chromatography on s ilica gel eluted by a hexane:ether (5:1) mixture (yields 50 55%). Undeca 1,10 dien 6 ol (4). 1 H NMR (300 MHz CDCl 3 ) 5.80 (m, 2H), 4.97 (m, 4H), 3.60 (br, 1H), 2.07 (q, 4H), 1.25 1.62 (br, 8 H). 13 C NMR (75 MHz, CDCl 3 ) 138.88, 114.75, 71.79, 37.06, 33.89 25.08. HRMS calcd. for C 11 H 20 O (M+H) + 169.1600; found, 169.1607. Anal. Calcd. for C 11 H 20 O: C, 78.51; H, 11.98; Found: C, 78.58; H, 12.03. Heptadeca 1,16 dien 9 ol (5). 1 H NMR (300 MHz CDCl 3 ) 5.81 (m, 2H), 4.97 (m, 4H), 3.58 (br, 1H), 2.06 (q, 4H), 1.2 2 1.53 (br, 20 H). 13 C NMR (75 MHz, CDCl 3 ) 139.33, 114.37, 72.18, 37.68, 33.98, 29.76, 29.31, 29.80, 25.82. HRMS calcd. for C 17 H 33 O (M+H) + 253.2531; found, 253.2534. Anal. Calcd. for C 17 H 33 O: C, 80.88; H, 12.78; Br. Found: C, 80.92; H, 12.83. Tricosa 1, 22 dien 12 ol (6). 1 H NMR (300 MHz CDCl 3 ) 5.82 (m, 2H), 4.98 (m, 4H), 3.59 (m, 1H), 2.04 (q, 4H), 1.20 1.50 (br, 32H). 13 C NMR (75 MHz, CDCl 3 ) 139.44, 114.30, 72.22, 37.7, 34.03, 29.92, 29.83, 29.77, 29.70, 29.35, 29.15, 25.87. HRMS calcd. for C 23 H 44 O (M) + 336.3314; found, 336.3306. Anal. Calcd. for C 23 H 44 O: C, 82.07; H, 13.18. Found: C, 82.15; H, 13.38. General procedure for bromination reaction. The precursor alcohols 4 5 and 6 were subjected to the same synthetic route which we reported earlier 168 to yield brominated ADMET monomers 6 bromoundeca 1,10 diene ( 7 ), 9 bromoheptadeca 1,16 diene ( 8 ), and 12
110 bromotricosa 1,22 diene ( 9 ). The brominated ADMET monomers were purified with column chromatography on s ilica gel eluted by hexane (yield 85%). 6 bromoundeca 1,10 diene (7). 1 H NMR (300 MHz CDCl 3 ) 5.80 (m, 2H), 5.00 (m, 4H), 4.05 (p, 1H), 2.05 (m, 4H), 1.82 (m, 4H), 1.77 1.41 (br, 4H). 13 C NMR (75 MHz, CDCl 3 ) 138.48, 115.14, 58.39, 38.72, 33.28, 26.98. H RMS calcd. for C 11 H 19 Br (M+Br) 380.9848; found, 380.9856. Anal. Calcd. for C 11 H 19 Br: C, 57.15; H, 8.28; Br, 34.56. Found: C, 57.24; H, 8.26; Br, 34.72. 9 bromoheptadeca 1,16 diene (8). 1 H NMR (300 MHz CDCl 3 ) 5.81 (m, 2H), 4.97 (m, 4H), 4.05 (p, 1H), 2 .05 (m, 4H), 1.80 (m, 4H), 1.61 1.30 (br, 16H). 13 C NMR (75 MHz, CDCl 3 ) 139.29, 114.45, 59.11, 39.37, 33.96, 29.17, 29.12, 29.04, 27.74. HRMS calcd. for C 17 H 31 Br (M + ), 314.1609; found, 314.1595. Anal. Calcd. for C 17 H 31 Br: C, 64.75; H, 9.91; Br, 25.34. Fo und: C, 64.97; H, 9.92; Br, 25.16. 12 bromotricosa 1,22 diene (9). 1 H NMR (300 MHz CDCl 3 ) 5.81 (m, 2H), 4.97 (m, 4H), 4.05 (p, 1H), 2.05 (m, 4H), 1.81 (m, 4H), 1.61 1.00 (br, 28H). 13 C NMR (75 MHz, CDCl 3 ) 139.44, 114.33, 59.23, 39.40, 34.03, 29.72, 29 .69, 29.67, 29.34, 29.28, 29.16, 27.80. HRMS calcd. for C 23 H 43 Br (M+Br) 417.1726; found, 417.1728. Anal. Calcd. for C 23 H 43 Br: C, 69.15; H, 10.85; Br, 20.00. Found: C, 69.16; H, 10.88; Br, 20.01. General procedure for bulk polymerization. Monomer and Gru catalyst were combined in a ratio of 500:1 under argon atmosphere. The polymerization was conducted at 35 40 C under vacuum with stirring for 5 days. The reaction was then stopped and 5 mL of toluene was added to dissolve the polymer with stirring. The reaction was allowed to cool to room temperature. The polymers were then precipitated by dripping the toluene solution into cold acidic methanol. They were then isolated by filtration and dried.
111 Polymerization of 6 bromoundeca 1,10 die ne (UPE9Br). 1 H NMR (300 MHz CDCl 3 ) 5.80 (b, 0.02H), (m, 2H), 5.00 (b, 0.06 H), 4.03 (p, 1H), 2.00 (m, 4H), 1.80 (m, 4H), 1.75 1.4 (b, 4 H). 13 C NMR (75 MHz, CDCl 3 ) 130.47, 129.96, 58.70, 58.63, 38.91, 38.83, 32.13, 27.83, 27.68, 26.81. M w (GPC vs. PS) = 18,400 g/mol. PDI = ( M w / M n ) = 1.99. Polymerization of 9 bromoheptadeca 1,16 diene (UPE15Br). 1 H NMR (300 MHz CDCl 3 ) 5.81 (b, 0.02H), (m, 2H), 4.97 (b, 0.04 H), 4.03 (p, 1H), 1.97 (m, 4H), 1.71 (m, 4H), 1.6 1.1 (b, 16 H). 13 C NMR (75 MHz, CDCl 3 ) 130.55, 130.07, 64.58, 38.76, 32.77, 29.88, 29.77, 29.38, 29.28, 29.24, 27.39, 26.71. M w (GPC vs. PS) = 35,300 g/mol. PDI = ( M w / M n ) = 1.75. Polymerization of 12 bromotricosa 1,22 diene (UPE21Br). 1 H NMR (300 MHz CDCl 3 ) 5.81 (b, 0.01H), (m, 2H), 5.00 (b, 0.02 H), 4.03 (p, 1H), 1.97 (m, 4H), 1.80 (m, 4H), 1.65 1.0 (b, 28 H). 13 C NMR (75 MHz, CDCl 3 ) 130.58, 130.12, 59.26, 39.63, 32.84, 29.89, 29.78, 29.53, 29.40, 29.32, 27.83, 27.44. M w (GPC vs. PS) = 75,800 g/mol. PDI = ( M w / M n ) = 2.19. General Procedure for hydrogenation. The bromine containing polymers ( UPE9Br ), ( UPE15Br ), and ( UPE21Br ) were hydrogenated using a 150 mL Parr high pressure reaction vessel equipped with a glass liner and Teflon stirbar Unsaturated polymer (1.0g) and glass liner under a nitrogen blanket. Finally, 20 mL of toluene were added. The vessel was sealed and attached to a grade 5 hydrogen tank and purged with hydrogen several times. The bomb was charged with 550 psi of H 2 and stirred for 5 days at room temper ature C. The hydrogenated polymer was dissolved in toluene, and precipitated into methanol. The polymer was then filtered and dried under reduced pressure. PE9Br. 1 H NMR (300 MHz CDCl 3 ) 4.03 (p, 1H), 1.81 (m, 4H), 1.65 1.2 (bm, 12H), 0.9 (t, 0.12H). 13 C NMR (75 MHz,CDCl 3 ) 59.23, 39.41, 29.62, 29.25, 27.80, 1.24. Anal. Calcd.: C,
112 52.70; H, 8.35; Br, 38.95. Found: C, 52.86; H, 8.36; Br, 38.72. M w (GPC vs. PS) = 23,600 g/mol. PDI = ( M w / M n ) = 1.75. PE15Br. 1 H NMR (300 MHz CDCl 3 ) 4.04 (p, 1H), 1.82 (m, 4H), 1.65 1.2 (bm, 24H) 0.9 (t, 0.20H). 13 C NMR (75 MHz,CDCl 3 ) 59.29, 39.42, 29.89, 29.87, 29.82, 29.74, 29.32, 27.83, 1.24. Anal. Calcd.: C, 62.28; H, 10.10; Br, 27.62. Found: C, 61.61; H, 10.09; Br, 27.26. M w (GPC vs. PS) = 27,600 g/mol. PDI = ( M w / M n ) = 1.68. PE21Br. 1 H NMR (300 MHz CDCl 3 ) 4.04 (p, 1H), 1.81 (m, 4H), 1.7 1.2 (bm, 36H), 0.9 (t, 0.11H). 13 C NMR (75 MHz,CDCl 3 ) 59.27, 39.42, 29.94, 29.88, 29.82, 29.74, 29.32, 27.82, 1.24. Anal. Calcd.: C, 67.54; H, 11.07; Br, 21.40. Found: C, 67.64; H, 11.06; Br, 21.39. M w (GPC vs. PS) = 94,100 g/mol. PDI = ( M w / M n ) = 2.23.
113 Table 4 1. Summary of thermal properties measured via DSC for the family of ethylene/vinyl bromide polymers. Sample M o (g/mol) H f (J/g) (b) H f (J/mol) (c) Tm( o C) Tm(K) S f (J \ molK) (d) S f \ bond (e) PE9Br 205.14 20.71 4248 14.81 258.35 16.44 1.83 PE15Br 289.29 35.05 10140 48.84 322.00 31.49 2.10 PE19Br (a) 345.40 43.35 14973 63.89 337.05 44.42 2.34 PE21Br 373.45 48.17 17989 69.63 342.79 52.48 2.50 (a) ref 168 The Tm and H values listed were obtained in a new specimen for baseline consistency within the series. (b), (c), (d), (e). From observed DSC values Figure 4 1. Synthesis of precisi on bromine polymers.
114 Figure 4 2. Thermogravimetric analysis and infrared spectra of PEXBr. (a) TGA for PE21Br (solid line), PE15Br (dashed line), and PE9Br (dotted line). (b) IR spectra for PE, PE21Br PE15Br and PE9Br
115 Figure 4 3. Melting and hea t of fusion trends for PEXBr polymers. (a) Relationship of the melting point (T m ) with the mole percentage of bromine in the repeating unit for PE, PE21Br (4.8% bromine), PE19Br (5.3% bromine), PE15Br (6.7% bromine) and PE9Br (11.1% bromine). (b) Relations hip of H f with the mole percentage of bromine in the repeating unit for PE, PE21Br (4.8% bromine), PE19Br (5.3% bromine), PE15Br (6.7% bromine) and PE9Br (11.1% bromine). The inset shows the second heating cycle for PE9Br
116 500 750 1000 1250 1500 1750 Wavenumber (cm-1)Transmittance (a. u.) UPE21Br UPE15Br UPE9Br Figure 4-4. IR spectra of UPE9Br, UPE15Br, UPE21Br. Indicat ed peaks correspond to those in Figure 4-2b, with the exception of the intense peak found at 967 cm-1 corresponding to the out of plane olefin C-H wag in the unhydrogenated samples UP21Br, UP15Br, and UP9Br shown here. 024681012 -20 0 20 40 60 80 100 120 140 PE11CH3PE19CH3PE15CH3PE21CH3PE9CH3PETm(oC)Mole % CH3 / Repeating Unit Figure 4-5. Tm vs number of moles of CH3 in repeating unit for PE9CH3, PE11CH3, PE15CH3, PE19CH3, PE21CH3, and PE.
117 CHAPTER 5 WELL DEFINED PRECISI ON ETHYL ENE/VINYL FLUORIDE P OLYMERS VIA CONDENSATION POLYMER IZATION 5.1 Introduction Fluoropolymers are industrially important materials that find application in areas where thermal and chemical resistance are critical to pe rformance. 190 As the first and most well known fluoropolymer, poly(tetrafluoroethylene) (PTFE or Teflon) finds numerous applications in household, industrial, and aerospace settings. 66 While PTFE is a fully fluor inated polyethylene (PE) analogue, numerous partially fluorinated analogues are also known and used commercially in a variety of applications. Among the most well known partially fluorinated polymers are poly(vinylfluoride) (PVF), poly(vinylidene fluoride) (PVDF), poly(ethylene tetrafluoroethylene) (PETFE), and random ethylene vinyl fluoride (EVF) polymers. 11 191 In all such polymers, a variation in properties is achieved via the tuning of the fluorine content as well as the distribution of hydrogens and fluorines in the backbone. 191 In an attempt to correlate properties with polymer composition in these partially fluorinated polymers, the presence of defects within t he polymer backbones must be taken into account. Common defects include head to head and tail to tail linkages and chain branching. 11 In order to fully correlate fluorine content with physical properties, a fam ily of well defined and defect free partially fluorinated polymers is required. Herein we present a family of precision EVF polymers with a fluorine atom on each and every 9 th 15 th and 21 st carbon, synthesized using acyclic diene metathesis (ADMET) polym erization followed by hydrogenation. In addition to providing the means of studying the properties of EVF polymers, characterization of this family of precision EVF polymers provides a comparison to the previously reported precision ethylene/vinyl chloride 192 (EVC) and precision ethylene vinyl/bromide 169 (EVB) polymers with
118 matched compositions. Therefore, a comparison based not only on halogen content, but based on halogen size is presented here. 5.2 Results and Discussion 5.2.1 Monomer and Polymer Synthesis Synthesis of the necessary fluorinated diene monomers is illustrated in Figure 5 1. The alcohol precursors (4 6) were prepared as described earlier. 169 Conversion of the alcohols to the corresponding fluorine monomers (7 9) was realized by react ion with DAST ( (Diethylamino)sulfur trifluoride). Polymerization was carried out in toluene at 45 C with diimide reduction 144 to give the polymers PE9F, PE15F, and PE21F. The nomenclature acronym used herein is PEXF, which indicates a polyethylene backbone with a fluorine substituent on every X th carbon, where X = 9, 15, 21 Polymer molecular weights are shown in Table 5 1 and vary between 7,600 and 10,400 g/mol. 5.2.2 Primary Structure Characterization The primary structure of these precise EVF polymers was established using a combination of 1 H NMR, 13 C NMR, elemental analysis, IR spectroscopy, and TGA (thermogravime tric analysis). The precise structure of the polymers is supported by NMR and the composition is confirmed by elemental analysis. The IR spectra in Figure 5 2 also provide characteristic information regarding the primary structure of the polymers, where th e absence of a peak at 967 cm 1 for all three EVF polymers indicates complete hydrogenation, based on the disappearance of the out of plane olefin C H wagging vibrational mode. 145 Characteristic peaks are found in several regions of the IR spectra. Vibrational modes associated with C F stretching are observed at 1069 cm 1 for all three EVF polymers. 147 The strong peaks at 1091 cm 1 and 1019 cm 1 in PE9F are due to a combination of C F and C C stretching. 193 The accentuation of these additional C F
119 related modes in PE9F is caused by the higher fluorine concentration relative to PE15F and PE21F. Doublets corresponding to methylene bending modes at ~1472 1463 cm 1 and meth ylene rocking modes at ~730 720 cm 1 are indicative of methylene sequences in PE analogues with an orthorhombic crystal structure. 194 195 Notice also a shoulder in all of the EVF samples at ~1434, which can be attributed to bending modes for the methylene gro ups to the C F groups. 148 5.2.3 Thermal and X Ray Analysis While NMR and IR support the proposed primary structure, TGA results directly support the precise composition of the polymers. Figure 5 3 displays the thermal decomposition curves of all three EVF polymers. In all cases, a two stage decomposition is observed, where the first stage corresponds to the loss of HF and the second stage marks the catastrophic decomposition of the polymer. Analogous to our previous work on ethylene vinyl halide polymers, the mass loss in the first stage quantitatively reflects the halogen content of the polymer. 168 The observed values for mass loss in the first stage are found to be in agreement with the calculated HF content for each of the EVF polymers at 14%, 9%, and 6% for PE9F, PE15F, and PE21F respectively. Figure 5 3 also shows that the onset of decomposition for the first stage loss of HF increases with decreasing fluorine content (Table 5 1). Therefore, EVF samples become more stable with increasing content of ethylene units as the labile fluorine content decreases. For comparison, the onset of decomposition for the first stage HF loss in PVF is 455 C. 196 The significantly higher value, relative to the EVF polymers reported here, can be attributed to the strong dipole dipole interactions present in PVF, which results in the enhanced stability. Figure 5 4 shows the second cycle of heating and cooling as measured by DSC for PE 9F, PE15F, and PE21F. Several key pieces of information are readily apparent. First, the melting temperatures (T m ) and crystallization temperatures (T c ) of the three polymers are essentially
120 equivalent (Table 5 1). Importantly, both melting and crystalliza tion are marked by sharp transitions, characteristic of the crystallization of homopolymers of similar molar mass, in which halogens are incorporated into the crystal at a composition equivalent to the overall composition of the polymer. This has been conf irmed by XRD and SSNMR previously for the case of PE19F. 168 From the heat of melting and crystallization the degrees of crystallinity were estimated under the assumption that the heat of fusion per mole of pure crystalline unit is, for the three EVF samples, the same as for polyethylene. The values, listed in Table 5 1, decrease from 74 to 53% with increasing fluorine content. WAXS patterns of precision EVF samples and a linear polyethylene fraction, shown in Fi gure 5 5, confirm that the orthorhombic packing of the un substituted chain is maintained in the EVF series. Clearly, the linear polyethylene and molecules with precise substitution of fluorine atoms at distances even as close as nine methylenes, are isomo rphous materials. Small shifts of the two main reflections to lower angles indicate some expansion of the unit cell axes due to the substitution of hydrogen for the larger fluorine atom. In addition, the tacticity can not be controlled in the ADMET synthes is, thus, the fluorine substitution imparts a defected nature to the precision molecules, that attests for the depression of the melting temperature from 133 C for the PE chain to 124 1 C, for the EVF series. Nonetheless, the expansion is not as severe a s that rendered by larger halogens (Cl, Br), which crystallize in a less symmetric triclinic packing. 168 Values of the unit cell dimensions, densities of the EVF orthorhombic lattices calculated assuming a unif orm fluorine distribution and constant c axis, and degrees of crystallinity calculated from the WAXD patterns after peak deconvolution, are listed in Table 5 2. The decrease in crystallinity with fluorine parallels the estimated values from heat of fusion and reflects, in reference to polyethylene, the perturbation of the fluorine atom to the
121 development of crystallinity. The similarity between both crystallinities supports the assumption of minor differences in H o (energy to melt one mole of crystalline r epeated backbone unit) in the EVF series. The DSC results point to an interesting comparison of EVF with our previous work on precision EVC 192 and ethylene vinyl bromide EVB 169 polymers of analogous halogen content. Figure 5 6a illustrates the variation in polymer melting temperature relative to halogen size and content. It is observed that for a constant mole percentage of halogen per repeating unit, the T m decrease s as the size of the halogen increases. This is in accord with our previously reported comparison of PE19X polymers (where X = F, Cl, and Br) 168 and is attributed to the accommodation of the halogens into the c rystal lattice, which serve as defects, lowering the T m by an amount proportional to the atomic size of the halogen. Notice also in Figure 5 6a, that for a constant halogen substituent, the T m decreases as halogen content increases for the case of Cl and B r, but not for F. It is clear that accommodation of greater contents of the larger Cl (van der Waals radius = 1.75 ) and Br (van der Waals radius = 1.85 ) atoms into the lattice serves as a defect, that changes the crystalline packing of the precision mo lecules with respect to the un substituted chain, thus resulting in a concomitant proportional decrease in T m However, the compositionally invariant T m for the fluorine containing polymers attests to the fact that the size of the fluorine atom (van der Wa als radius = 1.47 ) is sufficiently small relative to a hydrogen atom (van der Waals radius = 1.2 ), such that lattice strains due to the inclusion of fluorine are minimal and the orthorhombic packing is maintained. Figure 5 6b illustrates the variation in H f relative to halogen size and content for all of the precision EVH polymers. The first observation based on these relationships is that for a constant halogen composition per repeat unit, the H f decreases with increasing halogen size. This result is indicative of the amount of strain that each
122 halogen induces upon incorporation into the crystal lattice as it relates to the amount of energy that must be added to the system to effect melting. The other observation is that, for the same halogen atom, as the content of halogen increases, the H f decreases. This has already been explained for the case of the precision EVC 192 and EVB 169 polymers where the incorporation of more halogen atoms into the cryst als results in a decrease in crystallinity and in the amount of energy required to disorder the all trans backbone in the crystal lattice. Similarly, for the EVF polymers the percent crystallinity decreases with increasing fluorine content, as calculated f rom heat of fusion and WAXD. 5.3 Conclusions Herein we have described the synthesis of a family of precision EVF polymers, which have a F atom on each and every 9 th 15 th and 21 st carbon. The precise primary structures and compositions are supported b y NMR, IR, TGA, and elemental analysis. In a manner analogous to compositionally matched precision EVC and EVB polymers, DSC shows sharp melting and crystallization transitions characteristic of homopolymers. As opposed to these precision EVC and EVB polym ers with analogous compositions, which show decreasing T m values with increasing halogen content, the T m values of the EVF polymers are compositionally invariant due to the small size of the F atom. Thus, while all precision EVH polymers show the character istics of homopolymers, the size of the halogen has a significant effect on the specific thermal properties of the polymer. 5.4 Experimental Chemicals. Chemicals were purchased from the Aldrich Chemical Company and used as rst generation ruthenium catalyst, bis(tricyclohexylphosphine) benzylidine ruthenium (IV)dichloride, was purchased from Strem Chemical and stored in an Argon filled dry box prior to use. Methylene chloride and o xylene were distilled over CaH 2
123 Instrume ntation. Solution 1 H NMR (300 MHz) and 13 C NMR (75 MHz) spectra were recorded on Mercury 300 spectrometer. All chemical shifts for 1 H and 13 C NMR were referenced to residual signals from CDCl 3 ( 1 H = 7.27 ppm and 13 C = 77.23 ppm) and to residual signals fro m C 6 D 5 CD 3 ( 1 H = 2.09 ppm and 13 C = 137.86 ppm) with an internal reference TMS 0.03% v/v to internal TMS standard for 0. In all the NMR work the solvents were chloroform d or toluene d 8 the temperature was 25 C or 80 C. H igh resolution mass spectral (HR MS) data were obtained on a Finnegan 4500 gas chromatograph/mass spectrometer using the electron ionization (EI) mode. Elemental analyses were carried out by Atlantic Microlabs Inc., Norcross, GA. The GPC measurements for samples were taken on a Waters Ass ociates 150C high temperature gel permeation chromatograph equipped with three Polymer Laboratories mixed bed Type B columns and an internal DRI detector. The mobile phase was BHT inhibited 1,2,4 trichlorobenzene (135 C, flow rate 1.0 mL/minute, typical s ample concentration 2 mg/mL). IR data was obtained using a Perkin Elmer Spectrum One FTIR outfitted with a LiTaO 3 detector. Measurements were automatically corrected for water and carbon dioxide. Thermogravimetric analysis (TGA) data was obtained with a Pe rkin Elmer 7 series thermal analysis system. The TGA samples (2 5 mg) were heated from 50 C to 700 C at 10 C/min under nitrogen. Melting and crystallizations were obtained at 10 C/min in a differential scanning calorimeter TA Instrument DSC Q1000 V9.6 Build 290 under nitrogen flow and calibrated with indium. WAXD diffractograms were collected at room temperature on samples crystallized from the melt at ~ 1 C/min using a slit collimated Siemens D between 5 and 40 with a step size of 0.02. The instrument was calibrated for d spacing with a standard polished piece of polycrystalline quartz, and the film thickness was offset using shims.
124 Synthesis: General procedure for Grignard reaction. Synthesis of 5 bromopent 1 ene (1 ), 8 bromooct 1 ene (2), 11 bromoundec 1 ene (3) undeca 1,10 dien 6 ol (4), heptadeca 1,16 dien 9 ol (5), and tricosa 1,22 dien 12 ol (6) was describe previously. 169 General procedure for fluorination reaction A solution of diethylaminosulfur trifluoride (DAST) (2.0 equiv.) in CH 2 Cl 2 was cooled to 78 C and a solution of the precursor alcohol 4, 5, or 6 (1 equiv.) in CH 2 Cl 2 and dry pyridine (2.5 mL) was added dropwise. The mixture was stirred at this temperat ure for 2 h and then warmed to room temperature and stirred overnight. At this time water was added and the organic phase was extracted with CH 2 Cl 2 and then dried with Na 2 SO 4 and then the solvent was removed to give the colorless oils 6 fluoundeca 1,10 die ne (7) and 9 fluoroheptadeca 1,16 diene (8), and the white solid 12 fluorotricosa 1,22 diene (9) which were purified by chromatography using 97:3 hexanes:ethyl acetate to give (55 60%) of the product. 6 fluoroundeca 1,10 diene (7). 1 H NMR (300 MHz CDCl 3 ) 5.81 (m, 2H), 5.00 (m, 4H), 4.57 4,41 (dp, 1H), 2.09 (m, 4H), 1.78 1.40 (br, 8H). 13 C NMR (75 MHz, CDCl 3 ) 138.67, 114.99, 95.51, 93.30, 34.90, 34.63, 33.71, 24.61, 24.55. HRMS calcd. for C 11 H 19 F (M+H) + 170.1478; found, 170.1474. Anal. Calcd. for C 11 H 1 9 F: C, 77.59; H, 11.25; F, 11.16. Found: C, 77.62; H, 11.50; F, 11.01. 9 fluoroheptadeca 1,16 diene (8). 1 H NMR (300 MHz CDCl 3 ) 5.81 (m, 2H), 4.99 (m, 4H), 4.56 4.40 (dp, 1H), 2.06 (m, 4H), 1.69 1.20 (br, 20H). 13 C NMR (75 MHz, CDCl 3 ) 139.32, 114.42, 95.85, 93.64, 35.51, 35.24, 33.98, 29.58, 29.22, 29.05, 25.34, 25.28. HRMS calcd. for C 17 H 31 F (M + ), 254.2404; found, 254.2418. Anal. Calcd. for C 17 H 31 F: C, 80.25; H, 12.28; F, 7.47. Found: C, 80.15; H, 12.36; F, 7.41.
125 12 fluorotricosa 1,22 diene (9). 1 H N MR (300 MHz CDCl 3 ) 5.81 (m, 2H), 4.97 (m, 4H), 4.56 4.40 (dp, 1H), 2.06 (m, 4H), 1.78 1.20 (br, 32H). 13 C NMR (75 MHz, CDCl 3 ) 139.43, 114.32, 95.89, 93.68, 35.54, 35.27, 34.04, 29.74, 29.69, 29.35, 29.16, 25.39, 25.33. HRMS calcd. for C 23 H 43 F (M + ), 338.3343; found, 338.3 357. Anal. Calcd. for C 23 H 43 F: C, 81.59; H, 12.80; F, 5.61. Found: C, 81.59; H, 12.80; F, 5.43. General procedure for solution polymerization and hydrogenation. Monomer and gon and stirred at 45 C for 5 days. The same amount of catalyst (based on the above ratio) was added into the solution every 24 hours. After 5 days the reaction was then stopped and 50 mL of toluene was added to dissolve the polymer with stirring. The rea ction was allowed to cool to room temperature. The polymers were then precipitated by dripping the toluene solution into cold acidic methanol. They were then isolated by filtration and dried. The unsaturated fluorine containing polymers were then hydrogena ted using a modified version of the method described by Hahn 144 by dissolving in dry o xylene under argon and adding 3.3 equivalents of p toluenesulfonyl hydrazide (TSH) and 4 equivalents of tri n propyl amine (TPA). The solutions were refluxed for 9 hours and then cooled to room temperature. The hydrogenated polymer was precipitated into ice cold methanol and isolated by filtration. The dried polymer was then redissolved in toluene and re precipitated by drippi ng into ice cold acidic methanol. A white solid was collected by filtration and the polymers were isolated in quantitative yield. PE9F. 1 H NMR (300 MHz toluene d 8) 4.44 4.28 (dp, 1H), 1.75 1.2 (bm, 16H). 13 C NMR (75 MHz, toluene d 8) 95.41, 93.16, 36.27, 35.98, 30.32, 30.25, 26.01, 25.95. Anal.
126 Calcd.: C, 75.89; H, 12.10; F, 12.00. Found: C, 66.36; H, 11.05; F, 10.77. M w (GPC vs. PE) = 8,900 g/mol. PDI = ( M w / M n ) = 2.0. PE15F. 1 H NMR (300 MHz toluene d 8) 4.44 4.28 (dp, 1H), 1.75 1.2 (bm, 28H). 13 C NMR (75 MHz, toluene d 8) 95.41, 93.17, 36.28, 36.00, 30.48, 30.41, 30.38, 30.00, 26.04, 25.97. Anal. Calcd.: C, 78.88; H, 12.80; F, 8.32. Found: C, 74.14; H, 12.37; F, 7.14. M w (GPC vs. PE) = 10,400 g/mol. PDI = ( M w / M n ) = 2.2. PE21F. 1 H NMR (300 MHz toluene d 8) 4.44 4.28 (dp, 1H), 1.75 1.2 (bm, 40H). 13 C NMR (75 MHz, toluene d 8) 95.41, 93.16, 36.28, 36.00, 30.53, 30.41, 30.38, 26.04, 25.97. Anal. Calcd.: C, 80. 70; H, 13.22; F, 6.08. Found: C, 79.01; H, 13.17; F, 5.83. M w (GPC vs. PE) = 7,600 g/mol. PDI = ( M w / M n ) = 1.8.
1 27 Table 5 1. Molar mass and thermal data of precision EVF Sample M w x10 3 M w \ M n T m (C) T c (C) H f (J/g) ( X c ) b First stage onset of decomposition c (C) Second stage onset of decomposition d (C) PE21F 7.6 a 1.8 124 113 205 74 294 418 PE15F 10.4 a 2.2 124 114 174 64 257 410 PE9F 8.9 a 2.0 123 114 137 53 210 408 a.) GPC vs. PE in DCB b.) % crystallinity ( X c ) from heat of fusion in reference to the value of fully crystalline polyethylene, 293 J/g 197 corrected by fluorine content. c.) Recorded at first stage 5% total mass loss under nitrogen gas, 10 C/min. d.) Recorded at second stage 10% total mass loss under n itrogen gas, 10 C/min. Table 5 2. Orthorhombic lattice parameters and crystallinity from WAXD patterns Sample 2 (110) 2 (200) a () b () c () V() 3 density (g/cm 3 ) Xc (x ray) (a) PE 21.44 23.78 7.484 4.978 2.547 94.894 0.9798 0.88 PE21F 21.38 23.68 7. 515 4.989 2.547 95.495 1.0332 0.87 PE15F 21.32 23.62 7.534 5.003 2.547 96.013 1.0514 0.75 PE9F 21.22 23.50 7.572 5.026 2.547 96.929 1.0962 0.62 (a) Degree of crystallinity Figure 5 1. Synthesis of precision EVF polyme rs.
128 600 800 1000 1200 1400 1600 PE9F PE15F PE21FWavenumber (cm-1)Transmittance (a. u.)PE 967 cm-1No C=C-H Figure 5-2. Infrared spectra of PE, PE21F, PE15F, PE9F. 100200300400500600 0 20 40 60 80 100 2n+1 (n=3,6,9) x F 2n+2 x +xHF Weight (%)Temperature (oC)PE9F PE15F PE21F Figure 5-3. Thermogravimetric analysis results for PE21F, PE15F, and PE9F.
129 20406080100120140160 PE21F PE15F PE15F Endo Heat Flow (a. u.)Temperature (oC) PE9F Figure 5-4. Differential scanning calorimetry exotherms and endotherms for PE9Cl, PE15Cl, PE21Cl.
130 Figure 5 5. Wide angle X ray diffractograms of a linear polyethylene fraction (M w = 16,500; M w /M n = 1.26) and ADMET precision EVF samples slowly cooled from the melt at ~ 1 C/min. Th e peak at 2 = 17.9 in PE21F belongs to some impurity.
131 Figure 5 6. Trends for the variation in T m (a) and H f (b) vs. mole % halogen per repeating unit.
132 CHAPTER 6 STATISTICALLY RANDOM DEFECT FREE ETHYLE NE / VINYL HALIDE MO DEL COPOLYMERS VIA CONDE NSATION POLYMERIZATI ON 6.1 Introduction Polyolefins represent the largest class of industrially produced polymers and their importance is manifested in diverse applications ranging from packaging to biomaterials and electronics. 1 Even though polyethylene (PE) itself possesses a broad and dynamic property set, additional application specific variations can be introduce d through copolymerization with other vinyl monomers. 118 119 Specifically, copolymerization with polar vinyl monomers can be used to vary the mechanical and physica l properties of the polymer. 75 198 However, direct copolymerization via common free radical methods often leads to poorly defined polymers with limited compositional ranges based on reactivity differences betwee n ethylene and polar monomers as well as the numerous defects (e.g. branching) introduced through free radical techniques. 198 199 Therefore, a variety of other polymerization methods have been employed in an eff ort to model the properties of such copolymers. 129 145 Olefin metathesis has emerged as an attractive method for modeling ethylene copolymers containing polar group s, based on the functional group tolerance of the employed late transition metal catalysts and the fidelity of the reaction, which provides perfectly linear polymers, free of structural defects. 76 Using Acycl ic Diene Metathesis (ADMET) we have pursued halogen containing ethylene polymers, and we have capitalized on the ability of ADMET to produce precision polymers in which fluorine, chlorine, and bromine atoms are separated by a constant methylene sequence. 168 169 192 200 In such systems, we have observed unique crystallization and thermal behavior attributed to the homopolym er like behavior of these precise ethylene / vinyl halide (EVH) polymers. Here we extend these EVH polymers for use as effective models for industrial polymers via copolymerization of halogen containing dienes with 1,9 decadiene to yield
133 statistically random and defect free EVH polymers. Such polymers are more relevant as model systems for chain addition EVH polymers produced by free radical techniques based on the ability to attain continuous methyle ne sequences of significant length and broad distributions of length. Variation in halogen size and content allows us to more accurately derive structure property relationships in such polymers. The defect free nature of the polymer backbone allows the use of NMR analysis to prove the random nature of the polymers, while thermal analysis via differential scanning calorimetry (DSC) provides insight into the compositional dependence of the thermal properties in these random polymers relative to their precise analogues. 6.2 Results and Discussion 6.2.1 Polymer Synthesis The synthesis of the statistically random copolymers is shown in Figure 6 1. Copolymerization was carried out using a halogen containing monomer (abbreviated [X]) and 1,9 decadiene (abbreviated [H]). Synthesis of the fluorine, 200 chlorine, 1 92 and bromine 169 containing monomers has been previously reported. Cop olymerization was carried out using RUPE15X and RUPE21X Following ADMET polymerization, the fluorine and chlorine containing copolymers were exhaustively hydrogenated via diimide reduction, 144 while 169 was used for hydrogenation of the bromine containing polymers to yield the fully saturated copolymers RPE15X an d RPE21X In the acronyms, R indicates random, U indicates unsaturation, and PE15X and PE21X indicate random polyethylene backbones with halogen contents equivalent to the precise analogues with a halogen on each and every 15 th or 21 st carbon. Molecular w eight data are given for all six of the random copolymers in Table 6 1.
134 6.2.2 Primary Structure Characterization The primary structure of these random copolymers RUPE15X, RUPE21X, RPE15X, and RPE21X was established using a combination of 1 H NMR, 13 C NMR, elemental analysis, IR spectroscopy, and TGA (thermogravimetric analysis). In depth analysis by 1 H and 13 C NMR described below, definitively establishes the statistically random nature of the copolymers presented. The assignment of signals in the 1 H and 13 C NMR spectra is presented for the Cl random copolymers, RUPE15Cl and RUPE21Cl as an example. The numbering of the positions is given in Figure 6 2: plain numbers for the 6 chloroundeca 1,10 diene moiety (Cl moiety) and prime numbers for the 1,9 decadien e moiety (H moiety). We add a t or a c after this number to mark the trans or cis configuration of the closest double bond. If this double bond connects to a 1,9 decadiene moiety t becomes and c becomes The proton spectra in Figure 6 3 for RUPE15Cl (top) and RUPE21Cl (bottom) can be easily assigned based on the chemical shifts of the homopolymers PE9Cl and PE15Cl 192 For the case of RUPE15Cl (top), seven regions can be integrated separately: 5.50 5.30 (3. 25H, ), 3.95 3.80 (1H, 5 ), 2.10 1.90 (6.60H, ), 1.80 1.65 (4.02H, 4 ), 1.65 1.55 (2.16H, 3a ), 1.55 1.40 (2.02H, 3b ) and 1.40 1.20 (5.57H, ). The above integrals were used to calculate the H moiety : Cl moiety ratio ([H]:[Cl]) for RUPE15Cl to be 0.64:1. The ratios derived from 1 H NMR integration are given for all random polymers (X = Cl, F, and Br moieties) in Table 6 2. The alkene region of the 13 C NMR spectra for RUPE15Cl and RUPE21Cl (Figure 6 4 top and bottom, respectively) shows four major s ignals and four minor ones. The most intense carbon at 130.24 corresponds to a symmetric trans double bond joining two Cl moieties, in our nomenclature, 1t Indeed, in the GHMBC spectrum for RUPE15Cl (Figure 6 5) this carbon couples with the proton at 5.39 both over one and over two bonds. The next two most intense
135 carbons, at 131.06 and 129.51, carry the protons at 5.41 and 5.37 correspondingly, and each of these carbons displays a cross peak with the protons of the other, meaning that they belong to the a symmetric trans double bond joining an H and a Cl moiety. Of these two carbons, 129.51 couples with 1.46 and 1.59 (as does 130.24, Figure 6 6) therefore it is the one from the Cl moiety, 1t m cross peaks over one and two bonds with 5.38, therefore it is from a trans double bond joining two H moieties, The signals of 1t + and correspond to the XX, HX and HH diads respectively, therefore the integrals of these signals ca n be used to determine the degree of randomness (DR) in the copolymers. Due to the signal overlap, three integrals can be measured reliably, corresponding to 1t + and Therefore we define DR = 2(XX+HH)/HX/(r + 1/r), where XX+HH is the integra l for 1t + HX is the sum of the integrals for and and r is the ratio of the H and X monomers. For a random copolymer DR = 1, for an alternating polymer DR = 0, for a block copolymer DR = 2 indicate a random distribution for all of these copolymers. Each of the alkene signals in Figure 6 4 is accompanied by a minor signal, ca 20 30% of the major, about 0.5 ppm upfield. These signals belong to the corres ponding cis double bond, as demonstrated by their cross peak with 2.05 ( H2c ) and 2.01 ( ) (Figure 6 7). display a pattern of three lines, while all the other lines for C1 and display a pattern of four lines (visible for RUPE21Cl Fi gure 6 4). Based on their intensities, the lines can be assigned to the tetrads. For instance, of the four lines of the most intense corresponds to the tetrad Cl H Cl Cl, the least intense to H H Cl H, and the two of equal intensities to Cl H Cl H and H H Cl Cl. The intensities of the lines for C1t suggest that the center, most intense line corresponds to
136 Cl Cl Cl Cl + H Cl Cl H, while the outer lines of equal intensities correspond to the two different carbons in the H Cl Cl Cl tetrad. The C5 signal a t 64.0 displays a pattern of three lines, corresponding to the triads Cl Cl Cl, HCl Cl and H Cl H (visible for RUPE21Cl Figure 6 8). H5 couples with the C4 carbons at 38.05, 37.96 (Figure 6 9) and the C3 carbons 26.56, 26.52, 26.43, 26.38 shown in Figure 6 10 (Figure 6 11). Based on the relative intensity and chemical shifts difference, these signals were C4 were used to determine the trans : cis ratios given in Table 6 2. The signals for positions 2 and were identified by their coupling with the alkene protons (Figure 6 12). Based on their chemical shifts and relative intensities, the signals at 32.61, 32.58, 32.05, 32.03, 27.26, 27.22, 26.69 and 26.65 have been assigned to 2t 2c and correspondingly. The remaining eight signals, all in the region 29.0 29.8 ppm (Figure 6 10), correspond to and The signals at 29.75, 29.70, 29.64 and 29.57 couple with the alkene protons, therefore they correspond to Based on their pattern of relative intensities, they have been assigned as and correspondingly. The signals for position 29.22, 29.19, 29.09 and 29.05, display a different intensities pattern, respectively 1 : 3.3 : 3.3 : 13.1, consistent with their chemical shifts sensing the configuration of the two double bonds in the H moiety only, and not the nature, H or Cl of the neighboring moieties. The differences in chemical shifts are also consistent with trans cis over 3 and 4 bonds seen in homopolymers UPE15Cl and UPE21Cl These signals have been assigned respectively to and where the first
137 letter denotes the configu ration of the closest double bond, and the second letter denotes the configuration of the other double bond originating from the same monomer unit. The analysis of the F and Br random copolymers was done in a similar way. The 13 C chemical shifts are given in Table 6 3. In the F copolymer, the carbons in positions 3 4 and 5 displayed couplings with fluorine. Values for the coupling constants are also given in Table 6 3. Table 6 4 presents the differences in 13 C chemical shifts for the same position in the three copolymers. These data indicate that the halogen makes no difference in the chemical shifts in and For position the difference exists only when across the double bond there is an X monomer unit, i.e. for and The lar ger the halogen, the more deshielded this position becomes. Differences in positions 2 5 are the same for indicating the additivity of the effects of the halogen and all the other effects on 13 C chemical shifts. 13 C chemical shifts in positio n 1 reflect whether past the double bonds there is a H or an X unit. Table 6 5 presents the trans cis and are characteristic for an unsubstituted alkyl chain, 192 while values in positions 3 and 4 reflect the presence of the halogen. Differen ces t (Table 6 6) indicate that an X unit past a trans double bond produces The differences for cis bonds are 0.03 ppm lower. Smaller differences t c can be seen for positions 2,3, and and the differences in positions 2 and are the same, as it is true for the differences in positions 3 and Data in Tables 6 4 6 6 indicate that the 13 C chemical shifts in polymers of the type (=CH CH2 CH2 CH2 CHX CH2 CH2 CH2=)n(=CH CH2 CH2 CH2 CH2 CH2 CH2=)m can be calculated with a precision of +/ 0.02 ppm by adding the effect of the halogen and the effect
138 of the configuration of the double bond. The corresponding increments are given in Table 6 7. As an example, the chemical shift for C3c in the Cl copolymers is 29.64 3.21 0.07+0.15=26.51 vs 26.52. For in the Br copolymers, the chemical shift is 130.34+0.80=131.14 vs. 131.12. The IR spectra in Figure 6 13 also provide characteristic informa tion regarding the primary structure of the polymers. Here the RPE15X polymers are shown as representative examples. In each case, characteristic peaks support the presence of the expected halogens. For RPE15F a peak at 1068 cm 1 corresponds to C F stretc hing vibrations. 147 For RPE15Cl the peaks at 612 and 664 cm 1 correspond to C Cl stretching vibrations. 148 Finally, peaks at 537 and 615 cm 1 for RPE15Br correspon d to C Br stretching vibrations. 183 Characteristic peaks at ~720 cm 1 and ~1470 cm 1 which correspond to vibrational modes of methylene sequences in PE analogues are also observed for each of the samples. For RPE15F the doublets observed at 729 721 cm 1 and 1472 1463 cm 1 are the same as observed in orthorhombic crystalline PE. 194 195 Singlets observed at 722, 1467 cm 1 and 722, 1466 cm 1 for RPE15Cl and RPE21Cl respectively, indicate a change in crystal packing when compared to RPE15F 6.2.3 Thermal Analysis TGA Figure 6 14 shows the TGA data for all six random polymers. It can be observed that all six polymers u ndergo two step degradation processes, marked by the initial loss of HX (X = F, Cl, Br) and followed by catastrophic decomposition. 168 The first stage loss of HX is a good measure of primary structure compositi on. In all cases the mass loss corresponding to HX conforms closely to the theoretically calculated value based on the monomer feed ratios in each polymer. For RPE21Br and RPE15Br the first stage mass loss gave approximately 22 and 28% mass loss, respecti vely, in agreement with the calculated HBr loss. For RPE21Cl and RPE15Cl the first stage mass loss gave approximately 11 and 15% mass loss, respectively, in accord with
139 the calculated HCl loss. For RPE21F and RPE15F the first stage mass loss gave approxi mately 6 and 9% mass loss, respectively, in accord with the calculated HF loss. 6.2.4 DSC The thermal properties of these random copolymers have been further investigated by DSC. Here the most interesting results are observed when compared to their precis e analogues as illustrated in Figure 6 15. A tabular summary of the data is also given in Table 6 1. Figure 6 15a shows the melting transitions for RPE21Cl PE21Cl RPE15Cl and PE15Cl Several observations can be made in the comparison of the melting beha vior of the direct analogues. First, it can be seen, as previously reported, 192 that PE21Cl has a T m 18 C higher than PE15Cl due to lower incorporation of Cl into the crystalline lattice of the PE21Cl Accord ingly, these precise polymers behave as homopolymers and display sharp melting transitions. Notice with the random copolymers, that much broader melt transitions are observed. When comparing direct analogues (either RPE21Cl and PE21Cl or RPE15Cl and PE15Cl ) it can be seen that the random copolymer displays a higher T m than the precise polymer, despite the same overall composition. This feature was also observed for methyl branched precise vs. random analogues 145 and provides additional evidence in support of very different crystallization mechanisms for these two types of polymers. Polymers with precisely placed halogens crystallize as homopolymers, as evidenced in our earlier work. 168 However, despite the feasibility of halogen participation in the crystal lattice, the crystallization of random copolymers is led by an early selection of sequences with the least constraints to form ordered arrays. Primarily methylen e sequences are selected first, followed by sequences with higher halogen contents. This sequence selection leads to a broader melt, due to the variable composition of the crystalline phase with crystallites defined by a range of Cl contents. This broad m elt is a reflection of the changing composition of the melt phase, which is originally rich in Cl, but
140 becomes progressively closer to the composition of the overall copolymer. The peak melting temperature of the random copolymers is higher than that of th e precise polymers due to the presence of lower defect content crystallites generated via sequence selection. The final melting transition comparison to make is between RPE15Cl and RPE21Cl where a T m 17 C higher is observed for the latter. This is expect ed due to the higher concentration of Cl in RPE15Cl leading to more defected crystals and a melt richer in Cl. A more dramatic example of the differences in crystallization between precise and random analogues is given by the melting traces after isotherma l crystallization shown in Figure 6 16 for increasing crystallization times. Single sharp melting peaks along the kinetic process of PE21Cl crystallized at 73 C are characteristic of homopolymers. The double melting of isothermally crystallized RPE21Cl is a general feature of random ethylene copolymers with comonomers that are totally or partially excluded from the crystalline regions. 201 202 It is explained as a result of partitioning of crystallizable sequences. 201 Long crystalline sequences, primarily methylene based, are selected first and form the early crystalline structure that further melts at the highest temperatures in Figure 6 16b. The remaining crystallizable sequences form a second population of cryst allites with slower kinetics, richer in Cl and morphologically different from the first. The lower melting endotherms of Figure 6 16b are associated with this second population of crystallites. They are formed to a large extent from sequences pinned to t he initially formed crystallites. Note that while the area under the high melting peak is basically constant with time, the lower melting endotherm increases continuously, reflecting large differences in kinetics. At the isothermal crystallization temper ature of 76 C, the initial selection of primarily long methylene sequences is very fast. However, pinning of the remaining sequences to the first population of crystallites imposes constraints in the melt topology for gathering additional
141 sequences riche r in Cl with a concomitant decrease in kinetics as reflected in the melting behaviour. The Br containing polymers behave similarly (Figure 6 15b). It is observed that while the precise polymers show sharp, well defined melt transitions, the random analogu es show broad double melting transitions. As discussed for Cl copolymers, this is evidence of a different mechanism for crystallization between the precise polymers and the random copolymers. The lack of a well defined T m in RPE21Br and RPE15Br is evidence that the homopolymer like crystallization of their precise analogues is no longer operative as is expected to be replaced by a mechanism based on selection of long most crystallizable sequences as described above for the Cl polymers. In the case of the fl uorine containing polymers, large differences in melting behavior are not expected, based on the small size of the fluorine atom, which does not introduce a large strain on incorporation into the crystal lattice. 200 In Figure 6 15c, results are shown for RPE21F PE21F RPE15F and PE15F Several observations can be made in this case. First, it is clear that all fluorine containing random copolymers and precise polymers display sharp, well defined melt transition s. Secondly, the peak melting temperatures of all four polymers are very close (in the range of 120 127 C). Looking more closely, for a comparison of RPE21F and PE21F it is observed that the random copolymer melts 3 C higher than its precise analogue (1 27 C vs. 124 C). It is expected that the random polymer will display a higher T m than its precise analogue, which has been shown to display homopolymer like crystallization due to a mechanism of selection of long crystallizable sequences, which is expect ed to operate in such random polymers. The two T m values are suspected to be so close based on the small size of the F atom and its ability to enter into the crystalline lattice with a minimum of disruption. Note that despite
142 the low strain exerted to the lattice by the F atom, the T m of RPE15F is 7 C lower than the value of RPE21F in agreement to the sequence selection based crystallization behaviour of a random copolymer. The early formed crystallites of RPE15F are in equilibrium with a melt with a comp osition in F richer than comparable crystallites from RPE21F. Thus, RPE15F crystals melt at lower T m 6.3 Conclusion and Outlook Here we have shown that ADMET copolymerization of a halogen containing diene and 1,9 decadiene is a useful method for the synthesis of industrially relevant model PE copolymers. As evidenced by NMR studies on the unsaturated precursor polymers, a statistically random distribution of comonomers is achieved in a defect free perfectly un branched polymer backbone. Thermal analys is via DSC confirms the distinct difference between precise and random EVH polymers produced via ADMET. In the case of precise polymers, sharp and well defined melt transitions show a homopolymer like crystallization in which all halogens are equally parti tioned between crystalline and non crystalline regions, while the random copolymers show broad melting transitions that conform to a mechanism of crystallization based on the selection of long crystallizable sequences. Future work will focus on elucidating the crystal structures of the random polymers via x ray diffraction and developing a quantitative measure of the distribution of halogens between the crystalline and amorphous phases via solid state NMR. 6.4 Experimental Chemicals. Chemicals were purcha sed from the Aldrich Chemical company and used as benzylidene ruthenium (IV)dichloride, was purchased from Strem Chemical and stored in an
143 Argon filled dry box 3 ) 3 was purchased from Strem. Methylene chloride and o xylene were distilled over CaH 2 Instrumentation. Spectrometer, operating at 500 MHz for H1 and 125 MHz for C13, equipped with a 5 mm indirect detection probe, and with z axis gradients. The temperature was 25 C and the solvent chloroform d H1 and C13 chemical shifts were carefully referenced to internal TMS. H1 spectra were collected in one transient, with a relaxation delay of 10 s and an acquisition time of 5 s. Baseline correction was applied prior to integration. C13 spectra were collected in 60,000 transients, with zero relaxation delay and an acquisition time of 3 seconds, which produc ed a digital resolution of approximately 1 ppb/point. GHMBC spectra were collected with 4096 points over a spectral window of 2600 Hz in f2 and 4096 increments over a spectral window of 15000 Hz in f1 with 16 transients per increment. The relaxation dela y was 1 s. The experiment was optimized for a long range H1 1 H NMR (300 MHz) and 13 C NMR (75 MHz) spectra were recorded on Mercury 300 spectrometer. All chemical shifts for 1 H and 13 C NMR were refe renced to residual signals from CDCl 3 ( 1 H = 7.27 ppm and 13 C = 77.23 ppm) and to residual signals from C 6 D 5 CD 3 ( 1 H = 2.09 ppm and 13 C = 137.86 ppm) with an internal reference TMS 0.03% v/v to internal TMS standard for 0. In all the NMR work the solvents we re chloroform d or toluene d 8 the temperatures were 25 C or 80 C. Elemental analyses were carried out by Atlantic Microlabs Inc., Norcross, GA. The GPC measurements for samples in THF were taken on a Waters GPCV 2K instrument Samples were run with HPLC grade THF at 40 C on Water Styragel HR 5E columns relative to polystyrene standards Polymer molecular weights reported versus polyethylene standards were
144 measured using a Waters Associates 150C high temperature gel permeation chromatograph equipped with three Polymer Laboratories mixed bed Type B columns and an internal DRI detector. The mobile phase was BHT inhibited 1,2,4 trichlorobenzene (135 C, flow rate 1.0 mL / minute, typical sample concentration 2 mg / mL). IR data was obtained using a Perkin El mer Spectrum One FTIR outfitted with a LiTaO 3 detector. Measurements were automatically corrected for water and carbon dioxide. Thermogravimetric analysis (TGA) data was obtained with a Perkin Elmer 7 series thermal analysis system. The TGA samples (2 5 mg ) were heated from 50 C to 800 C at 10 C/min. Melting and crystallizations were obtained at 10 C/min in a differential scanning calorimeter TA Instrument DSC Q1000 V9.6 Build 290 under nitrogen flow and calibrated with indium. General procedure for bu lk polymerization. catalyst were combined in a ratio of 500:1 under argon atmosphere for chlorine and bromine containing polymers. The polymerization was conducted at 35 40 C under vacuum with stirring for 5 days. The reaction was then stopped and 50 mL of toluene was added to dissolve the polymer with stirring. The reaction was allowed to cool to room temperature. The polymers were then precipitated by dripping the toluene solution into cold acidic methanol. They were then isolated by filtration and dried. General procedure for solution polymerization. catalyst (500:1 ratio) were dissolved in toluene for fluorine containing polymers under argon and stirred at 45 C for 5 days. The same amount of catalyst was added into the solution every 24 hours. The same procedure as described above was used to isolate the polymer.
145 RUPE15F. Synthesized by the solution method as above using 0.624 g (3.66x10 3 mol) 6 fluoroundeca 1,10 diene, 200 0.376 g (2.72x10 3 mol) 1,9 decadiene, and 1.05x10 2 g (1.28x10 5 1 H NMR and 13 C NMR data see Table 6 2. RUPE21F. Synthesized by the solution method as above using 0.454 g (2.67x10 3 mol) 6 fluoroundeca 1,10 diene, 200 0.546 g (3.95x10 3 mol) 1,9 decadiene, and 1.09x10 2 g (1.32x10 5 1 H NMR and 13 C NMR data see Table 6 2. RUPE15Cl. Synthesized by the bulk method as above using 0.964 g (5.16x10 3 mol) 6 chloroundeca 1,10 diene, 192 0.536 g (3.87x10 3 mol) 1,9 decadiene, and 1.48x10 2 g (1.81x10 5 mol) 1 H NMR and 13 C NMR data see Table 6 2. RUPE21Cl. Synthesized by the bulk method as above using 0.711 g (3.81x10 3 mol) 6 chloroundeca 1,10 diene, 192 0 .789 g (5.71x10 3 mol) 1,9 decadiene, and 1.56x10 2 g (1.90x10 5 1 H NMR and 13 C NMR data see Table 6 2. RUPE15Br. Synthesized by the bulk method as above using 1.035 g (4.48x10 3 mol) 6 bromo undeca 1,10 diene, 169 0.465 g (3.36x10 3 mol) 1,9 decadiene, and 1.29x10 2 g (1.57x10 5 1 H NMR and 13 C NMR data see Table 6 2. RUPE21Br. Synthesi zed by the bulk method as above using 0.791 g (3.42x10 3 mol) 6 bromoundeca 1,10 diene, 169 0.709 g (5.13x10 3 mol) 1,9 decadiene, and 1.40x10 2 g (1.71x10 5
146 yield. For 1 H NMR and 13 C NMR data see Table 6 2. General Procedure for hydrogenation. The polymers containing F or Cl halogens were then hydrogenated using a modified version of the method described by Hahn 144 by dissolving in dry o xylene under argon and adding 3.3 equivalents of p toluenesulfonyl hydrazide (TSH) and 4 equivalents of tri n propyl amine (TPA). The solutions were refluxed for 9 hours and then cooled to room temperature. The hydrogenated polymer was precipitated into ice cold methanol and isolated by filtration. The dried polymer was then redissolved in toluene and re precipitated by dipping into ice cold acidic methanol. A white solid was collected by filtration and the polymers were isolated in quantitative yield. The polymers containing Br halogen were hydrogenated using a 150 mL Parr high pressure reaction vessel equipped with a glass liner and Teflon stirbar Unsaturated polymer liner under nitrogen blanket. 169 Finally, 20 mL of toluene were added. The vessel was sealed and attached to a grade 5 hydrogen tank and purged with hydrogen several times. The bomb was charged with 500 psi of H 2 and stirred for 5 days at room temperature. The hydrogenated polymer was dissolved in toluene, and precipitated into methanol. The polymer was then filtered and dried under reduced pressure. RPE15F. Hydrogenation was performed as above. 1 H NMR (300 MHz toluene d 8) 4.53 4.37 (dp, 1H), 1.8 1.23 (bm, 28H). 13 C NMR (75 MHz, toluene d 8) 95.40, 93.16, 36.27, 35.99, 30.53, 30.41, 30.38, 30.32, 30.26, 26.01, 25.95. Anal. Calcd.: C, 78.88; H, 12.80; F, 8.32. Found: C, 71.38; H, 12.16; F, 6.97. M w (GPC vs. PE ) = 7,900 g/mol. PDI = ( M w / M n ) = 2.5. RPE21F. Hydrogenation was performed as above. 1 H NMR (300 MHz toluene d 8) 4.45 4.28 (dp, 1H), 1.8 1.23 (bm, 40H). 13 C NMR (75 MHz, toluene d 8) 95.41, 93.16, 36.28, 36.00,
147 30.53, 30.38, 25.97. Anal. Calcd.: C, 80. 70; H, 13.22; F, 6.08. Found: C, 72.63; H, 12.51; F, 5.35. M w (GPC vs. PE) = 21,600 g/mol. PDI = ( M w / M n ) = 1.6. RPE15Cl. Hydrogenation was performed as above. 1 H NMR (300 MHz, CDCl 3 ) 3.89 (p, 1H), 1.69 (m, 4H), 1.60 1.20 (b, 24H). 13 C NMR (75 MHz, CDCl 3 ) 64.61, 38.76, 29.94, 29.82, 29.76, 29.65, 29.43, 29.38, 26.72. Anal. Calcd.: C, 73.58; H, 11.94; Cl, 14.48. Found: C, 69.75; H, 11.74; Cl, 13.52. M w (GPC vs. PS) = 73,600 g/mol. PDI = ( M w / M n ) = 2.1. RPE21Cl. Hydrogenation was performed as above. 1 H NMR (300 MHz, CDCl 3 ) 3.90 (p, 1H), 1.70 (m, 4H), 1.60 1.17 (b, 36H). 13 C NMR (75 MHz, CDCl 3 ) 64.61, 38.76, 29.95, 29.83, 29.77, 29.65, 29.44, 29.39, 26.75. Anal. Calcd.: C, 76.66; H, 12.56; Cl, 10.78. Found: C, 75.04; H, 12.55; Cl, 10.84. M w (GPC vs. PS) = 118,500 g/mol. PDI = ( M w / M n ) = 2.2. RPE15Br. Hydrogenation was performed as above. 1 H NMR (300 MHz CDCl 3 ) 4.03 (p, 1H), 1.81 (m, 4H), 1.70 1.00 (bm, 24H). 13 C NMR (75 MHz,CDCl 3 ) 59.24, 39.40, 29.93, 29.82, 29.73, 29.61, 29.31, 29.25, 27.79. Ana l. Calcd.: C, 62.28; H, 10.10; Br, 27.62. Found: C, 60.34; H, 9.85; Br, 29.71. M w (GPC vs. PS) = 72,100 g/mol. PDI = ( M w / M n ) = 1.8. RPE21Br. Hydrogenation was performed as above. 1 H NMR (300 MHz CDCl 3 ) 4.03 (p, 1H), 1.81 (m, 4H), 1.65 1.15 (bm, 36H). 1 3 C NMR (75 MHz,CDCl 3 ) 59.23, 39.41, 29.94, 29.82, 29.74, 29.62, 29.31, 29.25, 27.82. Anal. Calcd.: C, 67.54; H, 11.07; Br, 21.40. Found: C, 65.68; H, 10.92; Br, 23.65. M w (GPC vs. PS) = 55,500 g/mol. PDI = ( M w / M n ) = 1.7.
148 Table 6 1. Polymer propert ies of precise and random ADMET samples. Sample M w x10 3 M w \ M n T m (C) T c (C) H f (J/g) H c (J/g) RPE21F 21.6 1.6 127 119 157 170 RPE15F 7.9 2.5 120 107 151 163 RPE21Cl 118.5 2.2 88 73 78 86 RPE15Cl 73.6 2.1 71 56 68 76 RPE21Br 5 5.5 1.7 56 71 RPE15Br 72.1 1.8 36 45 Table 6 2. Monomer ratio, degree of randomness (DR) and trans : cis ratio in copolymers. copolymer F Cl Br Ratio H:X in RUPE15X 0.70 0.64 0.43 Ratio H:X in RUPE21X 1.40 1.28 1.04 (X+HH)/HX in RUPE15X 1.01 1.19 1.08 (X+HH)/HX in RUPE21X 1.08 0.99 1.06 DR in RUPE15X 0.95 1.08 0.78 DR in RUPE21X 1.02 0.96 1.06 Ratio trans:cis in RUPE15X 3.34 3.68 3.72 Ratio trans:cis in RUPE21X 3.29 3.46 2.82 Table 6 3. C13 chemical shifts in random copolymers. 1 2 3 4 5 1' 2' 3' 4' F ( 1 J C F ) 4.0 20.9 167.6 t 130.30 32.34 25.05 34.63 94.29 131.00 32.59 29.59 29.05 t' 129.65 32.36 25.08 34.63 94.29 130.34 32.61 29.64 29.05 c 129.78 26.99 25.20 34.74 94.29 130.49 27.25 29.71 29.19 c' 129.17 26.96 25.23 34. 74 94.29 129.88 27.22 29.75 29.19 Cl t 130.24 32.03 26.38 37.96 63.92 131.06 32.58 29.57 29.05 t' 129.51 32.05 26.43 37.96 63.92 130.32 32.61 29.64 29.05 c 129.72 26.69 26.52 38.05 63.92 130.56 27.26 29.70 29.19 c' 129.02 26.65 26.56 38.0 5 63.92 129.86 27.22 29.75 29.19 Br t 130.25 31.90 27.44 38.60 58.42 131.12 32.57 29.56 29.04 t' 129.46 31.92 27.50 38.60 58.42 130.32 32.60 29.63 29.04 c 129.72 26.57 27.60 38.68 58.42 130.61 27.26 29.69 29.18 c' 128.96 26.53 27.63 38.68 58.42 129.86 27.22 29.74 29.18
149 Table 6 4. Differences in C13 chemical shifts between the Cl, Br, and F copolymers. 1 2 3 4 5 1' 2' 3' 4' Cl t 0.06 0.31 1.33 3.33 30.37 0.06 0.01 0.02 0.00 t' 0.14 0.31 1.35 3.33 30.37 0.02 0 .00 0.00 0.00 c 0.06 0.30 1.32 3.31 30.37 0.07 0.01 0.01 0.00 c' 0.15 0.31 1.33 3.31 30.37 0.02 0.00 0.00 0.00 Br t 0.05 0.44 2.39 3.97 35.87 0.12 0.02 0.03 0.01 t' 0.19 0.44 2.42 3.97 35.87 0.02 0.01 0.01 0.01 c 0.06 0.42 2.40 3.94 35.87 0.12 0.01 0.02 0.01 c' 0.21 0.43 2.40 3.94 35.87 0.02 0.00 0.01 0.01 Table 6 5. Differences in C13 chemical shifts between the trans and cis configuration of the double bond in copolymers. 1 2 3 4 5 1' 2' 3' 4' F t c 0.52 5.35 0.15 0.11 0.00 0.51 5.34 0.12 0.14 t' c' 0.48 5.40 0.15 0.11 0.00 0.46 5.39 0.11 0.14 Cl t c 0.52 5.34 0.14 0.09 0.00 0.50 5.32 0.13 0.14 t' c' 0.49 5.40 0.13 0.09 0.00 0.46 5.39 0.11 0.14 Br t c 0.53 5.33 0.16 0.08 0.00 0.51 5.31 0.13 0.14 t' c' 0.50 5.39 0.13 0.08 0.00 0.46 5.38 0.11 0.14 Table 6 6. Differences in C13 chemical shifts t and c in copolymers. 1 2 3 4 5 1' 2' 3' 4' F t t' 0.65 0.02 0.03 0.00 0.00 0.66 0.02 0.05 0.00 c c' 0.61 0.03 0.03 0.00 0.00 0.61 0.03 0.04 0.00 Cl t t' 0.73 0.02 0.05 0.00 0.00 0.74 0.03 0.07 0.00 c c' 0.70 0.04 0.04 0.00 0.00 0.70 0.04 0.05 0.00 Br t t' 0.79 0.02 0.06 0.00 0.00 0.80 0.03 0.07 0.00 c c' 0.76 0.04 0.03 0.00 0.00 0.75 0.04 0.05 0.00
150 Table 6 7. Increments for the calculus of the C13 chemical shifts in copolymers. 1 2 3 4 5 1' 2' 3' 4' base 130.34 32.61 29.64 29.05 29.49 130.34 32.61 29.64 29.05 i ncr ement F 0.69 0.25 4.56 5.58 64.80 0.66 0.02 0.05 0.00 incr ement Cl 0.83 0.56 3.21 8.91 34.43 0.74 0.03 0.07 0.00 incr ement Br 0.88 0.69 2.14 9.55 28.93 0.80 0.03 0.07 0.00 incr ement cis 0.52 5.34 0 .15 0.09 0.00 0.51 5.32 0.13 0.14 Figure 6 1. Synthesis of random copolymers via ADMET. Figure 6 2. Repeat unit structure for RPEXCl polymers. Figure 6 3. 1 H NMR spectra for RUPE15Cl (top) and RUPE21Cl (bottom).
151 Figure 6 4. 13 C N MR for RUPE15Cl (top) and RUPE21Cl (bottom). Figure 6 5. Expansion (1) of the GHMBC spectrum of RUPE15Cl.
152 Figure 6 6. Expansion (2) of the GHMBC spectrum of RUPE15Cl Figure 6 7. Expansion (3) of the GHMBC spectrum of RUPE15Cl
153 Figure 6 8 13 C NMR for RUPE15Cl (top) and RUPE21Cl (bottom). Figure 6 9. Expansion (4) of the GHMBC spectrum of RUPE15Cl
154 Figure 6 10 13 C NMR for RUPE15Cl (top) and RUPE21Cl (bottom). Figure 6 11. Expansion (5) of the GHMBC spectrum of RUPE15Cl.
155 Figure 6-12. Expansion (6) of the GHMBC spectrum of RUPE15Cl. 500 750 1000 1250 1500 RPE15Br RPE15ClWavenumber (cm-1)Transmittance (a. u.)RPE15F Figure 6-13. IR spectra for RPE15F (top), RPE15Cl (middle), and RPE15Br (bottom).
156 100200300400500600 RPE15Br Weight (%)Temperature (oC)= 20 wt(%) RPE21Br RPE15Cl RPE21Cl RPE15F RPE21F Figure 6-14. TGA results for all six random polymers, RPE21X (solid lines) and RPE15X (dashed lines), where X = F (top), Cl (middle), and Br (bottom).
157 Fi gure 6 15 DSC comparison of random copolymers and precise analogues. (a ) RPE21Cl, RPE15Cl, PE21Cl, PE15Cl. (a) RPE21Br, RPE15Br, PE21Br, PE15Br. (a) RPE21F, RPE 15F, PE21F, PE15F.
158 Figure 6-16. Melting therm ograms of precise PE21Cl (a) and random copolymer analogue RPE21Cl (b), isothermally crystallized at 76 C and 73 C respectively for the times indicated.
159 LIST OF REFERENCES 1. Mathot, V.B.F.; Reynaers, H. In Handbook of Thermal Analysis and Calorimetry 1st ed.; Cheng, S.Z.D., Ed.; Elsevier: New York, 2002; Vol. 3, p 197. 2. Kattas, L.; Gastrock, F.; Levin, I. ; Cacciatore, A. In Modern Plastics Handbook Ha rper, C. A., Ed.; McGraw Hill: New York, 2000; p 4.1. 3. Peacock, A, J. Handbook of Polyethylene: Structures, Properties, and Applications Marcel Dekker: New York, 2000; p 375. 4. Miller, E. Introduction to Plastics and Composites: Mechanical Properties a nd Engineering Applications Marcel Dekker: New York, 1996; p 81. 5. Iwamichi, H.; Adachi, Y. Jpn. Patent 74,027,108, 1974. 6. Frielink, J. G. Brit. Patent 1,161,958, 1969. 7. Ebnesajjad, S. In Encyclopedia of Polymer Science and Technology Kroschwitz, J. I., Ed.; Wiley Interscience: New Jersey 2003; Vol. 4, p 438. 8. Ebnesajjad, S. In Encyclopedia of Polymer Science and Technology Kroschwitz, J. I., Ed.; Wiley Interscience: New Jersey 2003; Vol. 4, p 445 9. Scientific Polymer Products Inc., Ontario, NY. 10. Goerlitz, M.; Minke, R.; Trautvetter, W.; Weisgerber, G. Angew. Makromol. Chem. 1973 29 30 137 162. 11. Cais, R.E.; Kometani, J. M. Polymer 1988 29 168 172. 12. Zerbi, G.; Cortili, G. Spectrochim. Acta 1970 26 733 739. 13. Boyer, R. F. J. Po lym. Sci., Part C 1975 50 189 242. 14. Boyer, R. F.; Miller, R. L. Macromolecules 1977 10 1167 1169. 15. Natta, G.; Bassi, I. W.; Allegro, G. Atti Accad. Nazi. Lincei, Rend., Classe Sci. Fis., Mat. Nat. 1961 31, 850 856. 16. Lando, J. B.; Olf, H. G.; Peterlin, A. J. Polym. Sci. Polym. Chem. Ed. 1966 4 941 951. 17. Bank, M. I.; Krimm, S. J. Appl. Phys. 1968 39 4951 4958. 18. Kavesh, S.; Schultz, J. M. J. Polym. Sci., Polym. Phys. Ed. 1970 8 243 276. 19. Kitamaru, R.; Mandelkern, L. J. Polym. Sci. Polym. Phys. Ed. 1970 8 2079 2087.
160 20. Rosenberg, Y.; Siegmann, A.; Narkis, M.; Shkolnik, S. J. Appl. Polym. Sci. 1992 45 783 795. 21. Jennings, T. C., Starnes, W. H., Jr. In PVC Handbook 1st ed.; Wilkes, C. E., Summers, J. W., Daniels, C. A., Eds.; Hanser: Munich, 2005; p 95. 22. Witenhafer, D. E.; Poledna, D. J. In PVC Handbook 1st ed.; Wilkes, C. E., Summers, J. W., Daniels, C. A., Eds.; Hanser: Munich, 2005; p 57. 23. Carman, C. J. Macromolecules 1973 6 725 728. 24. Matthews, G. PVC: Production, Properties, and Uses The Institute of Materials: London, 1996; p 41. 25. Summers, J. W. J. Vinyl Technol. 1981 3 107 110. 26. Natta, G.; Corradini, P. J. Polym. Sci. 1956 20 251 266. 27. Witenhafer, D. E. J. Macromol. Sci. Phys. 1970 4 915 929. 28 Aubin, M.; Prud'homme, R. E. Macromolecules 1988 21 2945 2949. 29. Starnes, W. H. Prog. Polym. Sci. 2002 27 2133 2170. 30. Allsopp, M. W.; Vianello, G. In Encyclopedia of Polymer Science and Technology Kroschwitz, J. I., Ed.; Wiley Interscience: N ew Jersey 2003; Vol. 8, p 437. 31. Ramey, K. C.; Lini, D. C. In Encyclopedia of Polymer Science and Technology Mark, H. F.; Gaylord, N. G.; Bikales, N. M.O. Menges, G.; Kroschwitz J. I., Eds.; Wil ey Interscience: New York, 1971; Vol. 14, p 273. 32. Frata M.; Vidotto, G.; Talamini, G. Chimica e l'Industria 1966 48 42 44. 33. Riande E. S. E.; Delgado, M. P.; Barrales Rienda, J. M. Macromolecules 1982 15 1152 1157. 34. O'Shea, M. L.; Low, M. J. D.; Morterra, C. Mater. Chem. Phys. 1989 23 499 516. 35. Burnett, D. M.; Ross, F. L.; Hay, J. N. Polym. Lett. 1967 5 271 276. 36. Miller, R. L. In Polymer Handbook 4 th ed.; Brandrup, J.; Immergut, E. H.; Grulke, E. A. Eds.; Wiley Interscience: New York, 1999; Chapter 6. 37. Herman, J. A.; Roberge, P. C. Tran s. Farad. Soc. 1966 62 3183 3188. 38. Herman, J. A.; Roberge, P. C. Trans. Farad. Soc. 1969 65 1315 1324. 39. Dietrich, B.; Eckart, S. Kolloid Z. 1965 201 111 115.
161 40. Herman, J. A.; Roberge, P. C. J. Polym. Sci. 1962 62 S 116 S118. 41. Jungnickel B. J. In Polymeric Materials Encyclopedia Salamone, J. C. Ed.; CRC Press: Boca Raton, 1996; Vol. 9, p 7115. 42. Russo, S.; Behari, K.; Chengji, S.; Pianca, M.; Barchiesi, E.; Moggi, G. Polymer 1993 34 4777 4781. 43. Tashiro, K. In Ferroelectric Polyme rs: Chemistry, Physics, and Applications Nalwa, H., Ed.; Marcel Dekker: New York, 1995; p 97. 44. Maccone, P.; Brinati, G.; Arcella, V. Polym. Eng. Sci. 2000 40 761 767. 45. Hasegawa, R.; Kobayashi, M.; Tadokoro, H. Polym. J. 1972 3 591 599. 46. Hu mphrey, J. S.; Amin Sanayei, R. In Encyclopedia of Polymer Science and Technology Kroschwitz, J. I., Ed.; Wiley Interscience: New Jersey 2003; Vol. 4, p 510. 47. Bachmann, M. A.; Lando, J. B. Macromolecules 1981 14 40 46. 48. Lovinger, A. J. Macromole cules 1981 14 322 325. 49. Prest, W. M.; Luca, D. J. J. Appl.Phys. 1978 49 5042 5047. 50. Gregorio, R., Jr.; Cestari, M.; Chaves, N.; Nociti, P. S.; de Mendonca, J. A.; de Almeida Lucas, A. In Polymeric Materials Encyclopedia Salamone, J. C. Ed.; CRC Press: Boca Raton, 1996; Vol. 9, p 7128. 51. Liu, Z.; Marechal, P.; Jerome, R. Polymer 1997 38 4925 4929. 52. Lee, S.; Knaebel, K. S. J. Appl. Polym. Sci. 1997 64 455 492. 53. Kusanagi, H. Chem. Lett. 1997 7 683 684. 54. Wessling, R. A.; Gibbs, D. S .; Obi, B. E.; Beyer, D. E.; Delassus, P. T.; Howell, B. A. In Encyclopedia of Polymer Science and Technology Kroschwitz, J. I., Ed.; Wiley Interscience: New Jersey 2003; Vol. 4, p 458. 55. Kinsinger, J. B.; Fischer, T. M.; Wilson, C. W. III. J. Polym. S ci. Polym. Lett. Ed. 1967 5 285 294. 56. Matsuo, K.; Stockmayer, W. H. Macromolecules 1975 8 660 663. 57. Boyer, R. F.; Spencer, R. S. J. App. Phys. 1944 15 398 405. 58. Illers, K. H. Kolloid Z. 1963 190 16 34. 59. Landes, B. G.; De Lassus, P. T.; Harrison, I. R. J. Macromol. Sci. Phys. 1983 B22 735 745.
162 60. Landes, B. G.; Raich, W. J.; De Lassus, P. T.; Harrison, I. R. J. Macromol. Sci. Phys. 1983 B22 747 762. 61. Okuda, K J Polym. Sci., Part A: 1964 2 1749 1764. 62. Howell, B. A.; Delassus P. T. J. Polym. Sci., Part A: Polym. Chem. 1987 25 1697 1708. 63. Narita, S.; Ichinohe, S.; Enomoto, S. J. Polym. Sci. 1959 37 251 261. 64. Day, M.; Suprunchuk, T.; Cooney, J. D.; Wiles, D. M. J. Appl. Polym. Sci. 1987 33 2041 2052. 65. Narita, S.; Okuda, K. J. Polym. Sci. 1959 38 270 272. 66. Kerbow, D. L. In Polymeric Materials Encyclopedia Salamone, J. C. Ed.; CRC Press: Boca Raton, 1996; Vol. 9, p 6884. 67. Gangal, S. V. In Encyclopedia of Polymer Science and Technology Kroschwitz, J. I. Ed.; Wiley Interscience: New Jersey 2003; Vol. 3, p 378. 68. Chanzy, H. D.; Smith, P.; Revol, J. F. J. Polym. Sci., Part C: Polym. Lett. 1986 24 557 563. 69. Bunn, C. W.; Howells, E. R. J. Polym. Sci. 1955 18 307 310. 70. Bunn, C. W. J. Polym. Sci. 1955 16 323 343. 71. Khanna, Y. P.; Chomyn, G.; Kumar, R.; Murthy, N. S.; O'Brien, K. P.; Reimschuessel, A. C. Macromolecules 1990 23 2488 2494. 72. Scheerer, K.; Wilke, W. Colloid. Polym. Sci. 1987 265 206 209. 73. Ebnesajjad, S.; Khaladkar, P. R. In Fluoropolymers Applications in Chemical Processing Industries Ebnesajjad, S., Ed.; William Andrew: Norwich, 2005; p 15. 74. Bowmer, T. N.; Tonelli, A. E. Polymer 1985 26 1195 1201. 75. Boffa, L. S.; Novak, B. M. Chem. Rev. 2000 100 1479 1493. 76. Lehman S. E., Jr.; Wagener K. B.; Baugh, L. S.; Rucker, S. P.; Schulz, D. N.; Varma Nair, M.; Berluche, E. Macromolecules 2007 40, 2643 2656. 77. Boone, H. W.; Athey, P. S.; Mullins, M. J.; Philipp, D.; Muller, R.; Goddard, W. A. J. Am. Chem. Soc. 2002 1 24 8790 8791. 78. Stoeva, S.; Vlaev, L. Macromol. Chem. Phys. 2002 203, 346 353. 79. Chang, B. H.; Zeigler, R.; Hiltner, A. Polym. Eng. Sci. 1988 28, 1167 1172.
163 80. Era, V. A. Die Makromol. Chem. 1974 175 2191 2198. 81. Friese, K.; Hobelbarth, B.; Rei nhardt, J.; Newe, R. Die Angew. Makromol. Chem. 1996 234 119 132. 82. du Toit, F. J.; Sanderson, R. D.; Engelbrecht, W. J.; Wagener, J. B. J. Fluorine Chem. 1995 74 43 48. 83. Schonhorn, H.; Gallagher, P. K.; Luongo, J. P.; Padden, F. J., Jr. Macromole cules 1970 3 800 801. 84. Ferraria, A. M.; da Silva, J. D. L.; do Rego, A. M. B. Polymer 2003 44 7241 7249. 85. Ferraria, A. M.; da Silva, J. D. L.; do Rego, A. M. B. J. Fluorine Chem. 2004 125 1087 1094. 86. Cruz Barba, L. E.; Manolache, S.; Denes, F. Langmuir 2002 18 9393 9400. 87. Kharitonov a A. P.; Taege, R.; Ferrier, G.; Teplyakov, V.V.; Syrtsova, D.A.; Koops, G. H. J. Fluorine Chem 2005 126 251 263. 88. Stoeva, S.; Tsocheva, D.; Terlemezyan, L. J. Therm. Anal. Calorim. 2006 85 439 447. 8 9. Sobottka, J. Acta Polymerica 1983 34 647 650. 90. Zhikuan, C.; Lianghe, S.; Sheppard, R. N. Polymer 1984 25 369 374. 91. Saito, T.; Matsumura, Y.; Hayashi, S. Polym. J. 1970 1 639 655. 92. Hosselbarth, B. Die Angew. Makromol. Chem. 1995 231 161 169. 93. Chang, B. H.; Dai, J. W.; Siegmann, A.; Hiltner, A. Polym. Eng. Sci. 1988 28, 1173 1181. 94. Era, V. A.; Lindberg, J. J. J. Polym. Sci., Part:A2 1972 10 937 944. 95. Blundell, D. J.; Keller, A. J. Macromol. Sci., Part:B2 1968 2 337 359. 9 6. Grimm, H. J.; Thomas, E. L. Polymer 1985 26 38 44. 97. Keller, A.; Matreyek, W.; Winslow, F. H. J. Polym. Sci. 1962 62 291 300. 98. Stoeva, S.; Popov, A.; Rodriguez, R. Polymer 2004 45 6341 6348. 99. Cross, E. M.; McCarthy, T. J. Macromolecules 1 992 25 2603 2607. 100. Nakagawa, T.; Yamada, S. J. Appl. Polym. Sci. 1972 16 1997 2012. 101. Harrison, I. R.; Baer, E. J. Polym. Sci.: Part A 2 1971 9 1305 1324.
164 102. Arroyo, N. A.; Hiltner, A. J. Appl. Polym. Sci. 1979 23 1473 1485. 103. Landes, B G.; Harrison, I. R. Polymer 1987 28 911 917. 104. Jameison, F. A.; Schilling, F. C.; Tonelli, A. E. Macromolecules 1986 19 2168 2173. 105. Braun, D.; Mao, W.; Bohringer, B.; Garbella, R. W. Die Angew. Makromol. Chem. 1986 141 113 129. 106. Pourahma dy, N.; Bak, P. I.; Kinsey, R. A. J. Macromol. Sci. Pure Appl. Chem. 1992 29 959 974. 107. Cais, R. E.; Kometani, J. M. Macromolecules 1982 15 954 960. 108. Gomez, M. A.; Tonelli, A. E.; Lovinger, A. J.; Schilling, F. C.; Cozine, M. H.; Davis, D. D. Ma cromolecules 1989 22 4441 4451. 109. Stephens, C. H.; Yang, H.; Islam, M.; Chum, S. P.; Rowan, S. J.; Hiltner, A.; Baer, E. J. Polym. Sci.Part B: Polym. Phys. 2003 41 2062 2070. 110. Gutzler, F.; Wegner, G. Colloid and Polym. Sci. 1980 258 776 786. 111. Hanna, R. J.; Fields, J. W. J. Vinyl Tech. 1982 4 57 61. 112. Lopez, D.; Reinecke, H.; Hidalgo, M.; Mijangos, C. Polym. Int. 1997 44 1 10. 113. Baughman, T. W.; Wagener, K. B. Adv. Polym. Sci. 2005 176 1 42. 114. Watson, M. D.; Wagener, K. B. M acromolecules 2000 33 8963 8970. 115. Jama, C.; Quensierre, J. D.; Gengembre, L.; Moineau, V.; Grimblot, J.; Dessaux, O.; Goudmand, P. Surf. Interface Anal. 1999 27 653 658. 116. Peacock, A, J. Handbook of Polyethylene: Structures, Properties, and Appl ications Marcel Dekker: New York 2000; p 459 117. Anton, D. Adv. Mater. 1998 10 1197 1205. 118. Feldman, D.; Barbalata, A. Synthetic Polymers ; Chapman and Hall: London, 1996; Chapter 2. 119. Gottesman, R. T.; Goodman, D. In Applied Polymer Science 2 nd e d.; Tess, R. W.; Poehlein, G. W., Eds .; ACS: Washington, D. C., 1985; p 383 120. Mathews, G. PVC: Production, Properties, and Uses; The Institute of Materials: London 1996; p 280. 121. Yang, H.; Islam, M.; Budde, C.; Rowan, S. J. J. Polym. Sci., Part A : Polym. Chem. 2003 41 2107 2116.
165 122. Walsby, N.; Sundholm, F.; Kallio, T.; Sundholm, G. J. Polym. Sci., Part A: Polym. Chem. 2001 39 3008 3017. 123. Tonelli, A. E. Macromolecules 1982 15 290 293. 124. Endo, K.; Saitoh, M. Polym. Bull. 2003 49 4 11 416. 125. Schilling, F. C.; Tonelli, A. E.; Valenciano, M. Macromolecules 1985 19 356 360. 126. Stoeva, S. J. Appl. Polym. Sci. 2004 94 189 196. 127. Younkin, T. R.; Cornor, E. F.; Henderson, J. I.; Friedrich, S. K.; Grubbs, R. H.; Bansleben, D. A. Science 2000 287 460 462. 128. Smith, J. A.; Brzezinska, K. R.; Valenti, D. J.; Wagener, K. B. Macromolecules 2000 33 3781 3794. 129. Sworen, J. C.; Smith, J. A.; Berg, J. M.; Wagener, K. B. J. Am. Chem. Soc. 2004 126 11238 11246. 130. Alamo, R.; Dom szy, R.; Mandelkern, L. J. Phys. Chem. 1984 88 6587 6595. 131. Alamo, R.G., Mandelkern L. Thermochimica Acta 1994 238 155 201. 132. L. Mandelkern, Crystallization of Polymers 2 nd e d. ; Cambridge University Press: 2002; Vol. 1. 133. Isasi, J.R., Haigh, J.A., Graham, J.T., Mandelkern, L., Alamo. R.G. Polymer 2000 41 8813 8823. 134. Flory, P.J.; Trans. Faraday Soc. 1955 51 848 857. 135. Crist, B.; Howard, P.R. Macromolecules 1999 32 3057 3067 136. Chen, H.Y.; Chum, S.P.; Hiltner, A., Baer, E. J. P olym. Sci., Polym. Phys. Ed. 2001 39 1578 1593. 137. Ver Strate, G.; Wilchinsky, Z.W.; J. Polym. Sci., Part A 2 1971 9 127 142. 138. (a) Wild, L.; Blatz, C. I n New Advances in Polyolefins Young T.C. Ed.; Lenum Press: New York, 1993. (b) Anantawara skul, S.; Soares, J.B.P.; Wood Adams, P.M.; Monrabal, B. Polymer 2003 44 2393 2401. 139. Wang, W J.; Zhu, S. Macromolecules 2000 33 1157 1162. 140. Hu, W.; Sirota, E.B. Macromolecules 2003 36 5144 5149. 141. Ruiz de Ballesteros, O.; Auriemma, F.; Gue rra, G.; Corradini, P. Macromolecules 1996 29 7141 7148.
166 142. Hopkins, T. E.; Wagener, K. B. Macromolecules 2003 36 2206 2214. 143. Watson, M. D.; Wagener, K. B. Macromolecules 2000 33 5411 5417. 144. Hahn, S. F. J. Polym. Sci.: Part A: Polym. Chem. 1992 30 397 408. 145. Sworen, J. C.; Smith, J. A.; Wagener, K. B.; Baugh, L. S.; Rucker, S. P. J. Am. Chem. Soc. 2003 125 2228 2240. 146. Baughman, T. W.; van der AA, E.; Lehman, S. E.; Wagener, K. B. Macromolecules 2005 38 2550 2551. 147. du Toit, F. J.; Sanderson, R. D. J. Fluorine Chem. 1999 98 107 114. 148. Bowmer, T. N.; Tonelli, A. E., J. Polym. Sci.: Part B: Polym. Phys. 1986 24 1631 1650. 149. Hagemann, H.; Snyder, R.G.; Peacock, A.J.; Mandelkern, L. Macromolecules. 1989 22 3600 3606. 1 50. Gunter, L.; Wegner, G.; Smith, J. A.; Wagener K. B. Colloid. Polym. Sci. 2004 282 773 781. 151. Callister, W.D. Jr. Materials Science and Engineering. An Introduction Wiley: New York, 2003. 152. axis of the triclinic cell corresponds to two full repeating units (>40 ). 153. Simanke, A.G.; Alamo, R.G.; Galland, G. B.; Mauler, R. S. Macromolecules 2001 34 6959 6971. 154. VanderHart, D.L. J. Magnetic Reson. 1981 44 ,117 125. 155. VanderHart, D.L., Perez, E. Macromolecules 1986 19 1902 1909. 156. Alamo, R.G.; VanderHart, D.L.; Nyden, M.R.; Mandelkern, L. Macromolecules 2000 33 6094 6105. 157. Zhang, X.; Li, Z,; Yang, H.; Sun, C. Macromolecules 2004 37 73 93 7400. 158. Hosoda, S.; Nomura, H.; Gotoh, Y.; Kihara, H. Polymer 1990 31 1999 2005. 159. Holdsworth, Cutler D.J.; Hendra, P.J.; Cudby, M.E.A,; Willis, H.A. Polymer 1977 18 1005 1008. 160. Roe, R J.; Gieniewski, C. J. Crystal Growth 1980 48 295 302 161. Watson, M. D.; Wagener, K. B. Macromolecules 2000 33 3196 3201.
167 162. Braun, D. J. Vinyl Addit. Technol. 2001 7 168 176. 163. Smalley, D. In PVC Handbook Wilkes, C. E.; Summers, J. W.; Daniels, C. A., Eds.; Hanser: Munich, 2005 ; p679 164. Kline M. W.; Skiest, E. N. In Encyclopedia of PVC Nass, L. I., Ed.; Marcel Dekker: New York 1976; p 109 165. Kline, M. W.; Skiest, E. N. In Encyclopedia of PVC Nass, L. I., Ed.; Marcel Dekker: New York 1976; p 138 166. Wilkes, C. E.; Westfahl, J. C.; Bac kderf, R. H. J. Polym. Sci: Part A 1 1969 7 23 33. 167. Hagiwara, M.; Miura, T.; Kagiya, T. J. Polym. Sci: Part A 1 1969 7 513 523. 168. Boz, E.; Wagener, K. B.; Ghosal, A.; Fu, R.; Alamo, R. G. Macromolecules 2006 39 4437 4447. 169. Boz, E.; Nemeth, A. J.; Alamo, R. G.; Wagener, K. B. Adv. Synth. Catal. 2007 349 137 141. 170. Wu, T. K. Macromolecules 1973 6 737 741. 171. Vogl, O.; Qin, M. F.; Zilkha, A. Prog. Polym. Sci. 1999 24 1481 1525. 172. Stehling, F.C.; Mandelkern, L. Macromolecules 197 0 3 242 252. 173. Maynard, H. D.; Grubbs, R. H. Tetrahedron Lett. 1999 40 4137 4140. 174. Brandrup, J.; Immergut, E. H.; Grulke, E. A. Polymer Handbook 4 th ed., Wiley Interscience: New York, 1984. 175. G. Mathews, PVC: Production, Properties, and Uses The Institute of Materials : London, 1996 ; p 13. 176. Lukac, I. ; Pilichowski, J. F. ; Lacoste, J. Polym. Degrad. and Stab. 1998 61 79 85. 177. Chanunpanich, N. ; Ulman, A. ; Strzhemechny, Y. M. ; Schwarz, S. A. ; Janke, A. ; Braun, H. G. ; Kraztmuller, T. Lang muir 1999 15 2089 2094 178. Lukac, I. ; Pilichowski, J. F. ; Lacoste, J. Polym. Degrad. and Stab. 1998 61 343 347. 179. Frchet, J. M. J. J. Macromol. Sci. Chem. 1981 A15 877 890. 180. Janovic, Z. Polym. Degrad. and Stab. 1999 64 479 487. 181. Hopki ns, T. E. ; Wagener, K. B. Macromolecules 2004 37 1180 1189.
168 182. Mehl, M. ; Marongiu, A. ; Faravelli, T. ; Bozzano, G. ; Dente, M. ; Ranzi, E. J. Anal. Appl. Pyrolysis 2004 72 253 272. 183. Eguiluz, M. ; Ishida, H. ; Hiltner, A. J. Polym. Sci. Polym. Phys. 1 979 17 893 897. 184. Snyder, R. G. J Mol. Spectrosc. 1961 7 116 144. 185. Narita, S. ; Ichinohe, S. ; Enomoto, S. J. Polym. Sci. 1959 37 273 280. 186. Chanunpanich, N. ; Ulman, A. ; Malagon, A. ; Strzhemechny, Y. M. ; Schwarz, S. A.; Janke, A. ; Kraztmuller T. ; Braun, H. G. Langmuir 2000 16 3557 3560. 187. Natta, G. ; ; Bassi, I. W. ; Carella, G. Makromol. Chem. 1966 91 87 106. 188. Gianotti, G. ; Capizzi, A. Eur. Polym. J 1970 6 743 752. 189. Baughman, T. W. ; Sworen, J. C. ; Wagener, K. B. Tetrahedron 2004 60 10943 10948. 190. Scheirs, J. In Modern Fluoropolymers Scheirs, J., Ed .; Wiley: Chichester 1997; p 1. 191. Ameduri, B.; Boutevin, B. Well Architectured Fluoropolymers: Synthesis, Properties and Applications Elsevier: Amsterdam 2004 ; p 187. 192. Boz, E.; Nemeth, A. J.; Ghiviriga, I.; Jeon, K.; Alamo, R. G.; Wagener, K. B. Macromolecules accep ted. 193. Hong, J. W.; Lando, J. B.; Koenig, J. L.; Chough, S. H.; Krimm, S. Vibrational Spectroscopy 1992 3 55 66. 194. Tashiro, K.; Sasaki, S.; Kobayashi, M. Macromolecules 1996 29 7460 7469. 195. Ungar, G.; Zeng, X. B. Chem. Rev. 2001 101 4157 4188. 196. Impallomeni, G.; Montaudo, G.; Puglisi, C.; Scamporrino, E.; Vitalini, D. J. Appl. Polym. Sci. 1986 31 1269 1274. 197. Wunderlich, B Macromolecular Physics: Crystal Structure, Morphology, Defects Academic Press: New York, 1973; Vol. 1 p 401 198. Doak, K. W. In Encyclopedia of Polymer Science & Engineering 2 nd e d.; Mark, H. F. ; Bikales, N. M.; Overberger, C. G.; Menges, G.; Kroschw itz, J. I.; Eds.; Wiley: New York, 1986; Vol. 6, p 386. 199. Klimesch, R.; Littmann, D.; MaÂ¨hling, F. O In Encyclopedia of Materials: Science and Technology Buschow, K. H. J.; Cahn, R. W.; Flemings, M. C.; Ilschner, B.; Kramer, E. J.; Mahajan, S., Eds.; Else vier Science: New York, 2004; p 7181. 200. Boz, E.; Nemeth, A. J.; Jeon, K.; Alamo R. G. ; Wagener, K. B. Macromolecules submitted.
169 201. Crist, B.; Claudio, E.S. Macromolecules 1999 32 8945 9851. 202. Rabiej, S.; Goderis, B.; Janicki, J.; Mathot, V. B.F.; Koch, M.H.J.; Groeninckx, G.; Reynaers, H.; Gelan, J. Wlochowicz, A. Polymer 2004 45 8761 8778.
170 BIOGRAPHICAL SKETCH Emine Boz was born in Antalya, Turkey in 1978. She received her B.S. in Chemistry from Hacettepe University in Ankara, Turkey in 2001. She then went on to complete her M.Sc. in Inorganic Chemistry in 2003 at Hacettepe University. Emine then joined the research group of Prof. Kenneth B. Wagener at the University of Florida in 2003, where she qualified in the organic division and conducted studies on the synthesis and crystallization of halogen containing polyolefins.