Thermoplastic Elastome

Well-defined Thermoplastic Elastomers
Reversible networks based on hydrogen bonding
Well-defined Thermoplastic Elastomers

Reversible networks based on hydrogen bonding
PROEFSCHRIFT
ter verkrijging van de graad van doctor aan de Technische Universiteit Eindhoven, op gezag van de Rector Magnificus, prof.dr. R.A. van Santen, voor een commissie aangewezen door het College voor Promoties in het openbaar te verdedigen op donderdag 30 januari 2003 om 16.00 uur
door
Ronny Mathieu Versteegen

geboren te Tegelen
Dit proefschrift is goedgekeurd door de promotoren: prof.dr. E.W. Meijer en prof.dr. C.A. Hunter
Copromotor: dr. R.P. Sijbesma
This research has been financially supported by the Netherlands Organization for Scientific Research (NWO/CW). Omslag: Ron Versteegen, Paul Verspaget Druk: Universiteitsdrukkerij, Technische Universiteit Eindhoven
CIP-DATA LIBRARY TECHNISCHE UNIVERSITEIT EINDHOVEN
Versteegen, Ronny M. Well-defined thermoplastic elastomers : reversible networks based on hydrogen bonding / by Ronny M. Versteegen. Eindhoven : Technische Universiteit Eindhoven, 2003. Proefschrift. ISBN 90-386-2834-X NUR 913 Trefwoorden: supramoleculaire chemie; waterstofbruggen / polymeernetwerken; dendrimeren / polymeerstructuur / thermoplastische rubbers; polyurethanes Subject headings: supramolecular chemistry; hydrogen bond / polymer networks; dendrimers / polymer morphology / thermoplastic rubber; polyurethanes
Table of Contents 1
Thermoplastic elastomers and well-defined polymers 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 Thermoplastic elastomers Polyurethanes Novel isocyanate chemistry Spider dragline silk Well-defined thermoplastic elastomeric polymers Aim of this thesis Outline of this thesis References and notes 1 2 3 5 7 9 11 12 13
2
[n]-Polyurethanes and hyperbranched polyurethanes 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 2.9 Introduction Synthesis of -amino--alcohols Preparation of [n]-polyurethanes Thermal properties of [n]-polyurethanes Synthesis of hyperbranched polyurethanes Degree of branching of hyperbranched polyurethanes Conclusions Experimental section References and notes 15 16 18 19 22 24 27 29 30 36
3
Synthesis and characterization of block copoly(ether urea)s 3.1 Introduction 3.1.1 Thermoplastic elastomers 3.1.2 Urea-based TPEs Molecular design of block copoly(ether urea)s Synthesis of amine-terminated prepolymers 39 40 40 42 43 44
3.2 3.3
3.4 3.5
3.6 3.7 3.8 3.9
Synthesis of heterofunctional pTHF Preparation of block copoly(ether urea)s 3.5.1 Segmented copolymers with one urea group in the hard block 3.5.2 Segmented copolymers with two urea groups in the hard block 3.5.3 Segmented copolymers with three urea groups in the hard block 3.5.4 Segmented copolymers with four urea groups in the hard block 3.5.5 Segmented copolymers with a polydisperse hard block Hydrogen bonding within block copoly(ether urea)s Conclusions Experimental section References and notes
47 50 50 50 52 54 54 55 58 59 63
4
Properties and morphology of block copoly(ether urea)s 4.1 4.2 4.3 4.4 4.5 4.6 4.7 4.8 4.9 Introduction Crystal structure of model bisurea Thermal properties Morphology Mechanical properties Orientation during tensile testing Conclusions Experimental section References and notes 65 66 67 69 74 80 85 90 91 93
5
A modular approach to polymer modification and functionalization 5.1 5.2 5.3 5.4 5.5 5.6 Introduction Increase of hard block content Incorporation of dye molecules Conclusions Experimental section References and notes 95 96 97 99 103 104 106
6
Block copolymers containing a monodisperse soft block 6.1 6.2 6.3 6.4 6.5 6.6 6.7 Introduction Synthesis of monodisperse pTHF oligomers Synthesis of a block copolymer containing a monodisperse soft block Properties Conclusions and outlook Experimental section References and notes 107 108 110 114 115 118 119 121
7
Reversible networks based on dendrimer telechelic association 7.1 7.2 7.3 7.4 Introduction Synthesis of host and guests Complexation of polymeric guests and dendritic host Viscometry of polymeric guest dendritic host complexes 7.4.1 Polydisperse, bifunctional guests 7.4.2 Polydisperse, monofunctional guest 7.4.3 Monodisperse, bifunctional guest 7.4.4 Temperature-dependence Dynamic light scattering Conclusions and outlook Experimental sections References and notes 123 124 127 129 131 132 134 135 136 137 140 142 143 145 147 149 151
7.5 7.6 7.7 7.8
Summary Samenvatting Dankwoord Curriculum Vitae
1
Thermoplastic elastomers and well-defined polymers
Abstract This introductory chapter reviews some aspects of thermoplastic elastomers and well-defined polymers. Two types of thermoplastic elastomers are described in more detail: spandex, a highly elastic synthetic polyurethane, and spider dragline silk, an exceptionally tough biopolymer. The relationship between the morphologies of both materials and their properties is recognized, and this demonstrates the relevance of well-defined molecular architectures. These considerations served to formulate the aim and outline of this thesis.
Chapter 1
1.1 Thermoplastic elastomers Since the 1940s, many industrial research groups from all over the world have been looking for materials that can be used as an alternative for natural rubber. This led to the discovery of several types of thermoplastic elastomers (TPEs).1 TPEs possess many of the physical properties of rubbers, i.e. softness, flexibility, and elasticity; but in contrast to conventional rubbers, they can be processed as thermoplastic materials.2 Natural rubbers are covalently (irreversibly) crosslinked and, therefore, they cannot be processed once the material is shaped and vulcanized. In contrast, the crosslinks in thermoplastic elastomers are reversible in nature, allowing a processing applying conventional techniques for thermoplastics, e.g. injection molding and extrusion. Another advantage is the possibility to recycle scrap. The thermoreversible nature of the crosslinks in TPEs can be various: e.g. phase separation, crystallization, reversible chemical bonding3, hydrogen bonding, metal complexation, or electrostatic interactions.4 TPEs can be roughly divided into three main groups.5 The largest group of TPEs consists of the ABA triblock copolymers, like polystyrene/polydiene (SBS) block copolymers. In these materials, the crosslinking is due to the formation of glassy domains of polystyrene, that are embedded in the amorphous and rubbery polydiene phase.6 The second group consists of thermoplastic polyolefins blended with an elastomer, e.g. a blend of polypropylene and EPDM rubber. The third group concerns the segmented copolymers. These (multi)block copolymers consist of alternating hard and soft blocks (figure 1.1), and in this thesis, we will mainly focus on this type of TPEs.
hard soft segment segment
Figure 1.1: Schematic representation of a segmented copolymer.1 In the segmented copolymers, the hard blocks account for the mechanical stability of the material, since they give rise to reversible crosslinks, which are embedded in an amorphous phase with a low glass transition temperature. This phase is mainly composed of the soft segment, and gives the material its flexibility. In many TPEs, the reversible crosslinks are formed by crystalline or glassy domains. Above the melting point or glass transition temperature of these domains, a viscous polymer melt is obtained, that can be processed easily. Next to reversible crosslinks, the hard blocks also act as reinforcing fillers, causing the material to be stiffer at higher hard block content.
The hard blocks are generally based on polyester1,7, polyamide8, or polyurethane9 segments. At ambient temperatures, the hard blocks are incompatible with the soft blocks. This induces microphase separation by crystallization or liquidliquid demixing. A schematic representation of the morphology of segmented copolymers, as proposed by Cella, is depicted in figure 1.2.10 The ordered arrays are formed by crystallized hard blocks. Hard blocks that are not crystallized are dissolved in the amorphous, soft phase, and this incomplete phase separation leads to an increase of the glass transition temperature of the soft phase, which is undesired for the low temperature flexibility and elasticity of the material.
Figure 1.2: Schematic representation of the morphology of a segmented block copolymer: () hard block, (~~~) soft block. Upon deformation of these materials, irreversible changes in their morphologies may occur. Above the yield point, the hard blocks are disrupted and will reorganize, resulting in energy dissipation and incomplete recovery to their initial dimensions. This hysteresis leads to irreversible changes in the polymer properties, i.e. the Youngs modulus.
1.2 Polyurethanes From the segmented copolymers, the thermoplastic elastomeric polyurethanes (TPUs) are the most well-known. Polyurethanes form a versatile class of polymers, which are used in a broad range of applications2,11, like foams, coatings, fibers, and as biomedical materials12, but also in less down-to-earth applications like the landing pads of the lunar module.13 In fact, polyurethanes are a collective term for all polymers comprising urethane, urea, or other isocyanate derived groups, even if they are only a minor part of the total structure. Both urethane and urea groups are known for their ability to self-associate via hydrogen bonding (figure 1.3).14 The urethane group forms linear hydrogen bonds, and since it is asymmetric, it can do so in either a parallel or an antiparallel fashion. The urea group is symmetric, and forms bifurcated hydrogen bonds, causing the hydrogen bond strength of 3
Chapter 1
ureas to exceed that of urethanes. In principle, urethane and urea groups can form infinite stacks of hydrogen bonded arrays.
O N H O N H O N H O N H O N H O N H
Figure 1.3: Hydrogen bonding between urethane (left) and urea groups (right). The properties of polyurethanes depend strongly on their macromolecular structure, i.e. the nature and functionality of their constituting (macro)monomers. Covalently crosslinked networks are thermosets, and are used as both rigid and flexible foams; segmented polyurethanes show thermoplastic elastomeric properties, and are used e.g. in textile fibers, shoe soles, or hoses; linear homopolyurethanes are thermoplastic materials, and can be used in fibers15 or as biodegradable material.16 The world consumption of polyurethanes in 2000 amounted to about 8 million tons, with a global growth averaging around 3 4% a year.2 Pioneering work in the field of polyurethanes was done by O. Bayer since 1937.15 He described the polyaddition of diisocyanates and diols, yielding [m,n]-polyurethanes. Fibers of this polymer, named Perlon-U, were expected to compete with the nylons developed by DuPont several years earlier.17 However, due to their limited thermal stability and inferior mechanical properties, they are no longer of industrial importance. Soon, it was discovered that the performance of the polyurethanes improved dramatically upon replacing the shortchain diol by a long-chain diol, yielding highly elastic fibers. The thermoplastic elastomeric polyurethanes (TPUs) were soon produced on an industrial scale by a number of companies.1 The TPU consists basically of three building blocks (figure 1.4): 1) a long-chain diol, normally with a polyether or polyester backbone, 2) a diisocyanate, mostly an aromatic one, 3) a chain extender, such as a short-chain diol, or a diamine.
residue of long chain diol residue of chain extender residue of diisocyanate urethane group
Figure 1.4: Schematic representation of a TPU and its building blocks.
They are often prepared in a one pot procedure, in which the long-chain diol is first reacted with an excess of the diisocyanate, to form an isocyanate functionalized prepolymer. The latter is subsequently reacted with the chain extender which results in the formation of the high molecular weight polyurethane. When a diamine is used as chain extender, the TPU also contains urea groups, that increases the melting temperature of the hard block. The urethane (and/or urea) groups in the hard block form hydrogen bonds, which improve the association and crystallization. However, also TPUs containing secondary urethane groups, thus unable of forming hydrogen bonds, have been prepared.18 These polymers showed properties comparable with their hydrogen bonding analogues, since the crystallization of the hard block is not prevented by blocking the hydrogen bonding sites. One of the most interesting TPU fibers is spandex (Elastane, Lycra), which is produced by DuPont. This fiber is employed extensively in ladies garment, and is now expanding into all areas of clothing at an incredible rate.19 The macromolecular structure of this highly elastic material allows the fiber to stretch up to 600% and then relax to its original shape. The structure of one type of spandex is depicted below.20 It is prepared by reacting a prepolymer diol with an aromatic diisocyanate, in this case, methylene diphenylene isocyanate (MDI).3 In the second step, the chain extension, the macrodiisocyanate is reacted with a diamine in a highly polar solvent (e.g. dimethyl acetamide) to give a high molecular weight poly(ether urethane urea). The most frequently used short-chain aliphatic diamines are 1,2-ethylene diamine, 1,2-propylene diamine, or hydrazine. Due to the presence of urea groups, the melting point of the hard blocks is too high to allow melt spinning of the fibers. Therefore, the more expensive dry spinning procedure is used, in which a viscous solution of the polymer in a volatile solvent is spun in hot air, allowing the evaporation of the solvent. Some types of spandex are known to be crosslinked covalently during processing.
O CH2 CH2 O x N H CH2 N H O N N H H O N H CH2 N H O O n
spandex
1.3 Novel isocyanate chemistry Isocyanates, the primary building blocks of polyurethanes, can be prepared via several The only route that is practiced on a significant industrial scale is phosgenation of primary amines or their salts (scheme 1.1). The high toxicity of phosgene is only one of its disadvantages; the rather high temperature necessary to decompose the intermediate routes.
13,21
Chapter 1
carbamoyl chloride, and the poor selectivity toward different nucleophiles are some other drawbacks that limit the synthetic use of phosgene.
O R NH2
+
O Cl - HCl R N H Cl - HCl R N=C=O
Cl
Scheme 1.1: Preparation of isocyanates via phosgenation of amines. Some other methods which are used on laboratory scale include the Curtius rearrangement of acylazides, the dissociation of blocked isocyanates, the conversion of alkyl halides with alkali cyanates, and the dehydration of carbamic acids (adducts of amines and carbon dioxide).22 Recently, Peerlings reported on the potential of di-tert-butyl tricarbonate 1 as a versatile and mild reagent for the synthesis of isocyanates.23 Di-tert-butyl tricarbonate 1 is synthesized from carbon dioxide, potassium tert-butoxide, and phosgene,24 and is readily obtained in multigram quantities. Upon reaction of aliphatic or some aromatic amines with a stoichiometric amount of 1, a fast reaction occurs, in which the amine functionality is converted to an isocyanate within a few minutes at room temperature via a cascade of reactions (scheme 1.2). The last step of this cascade is the rate determining step, and since this step is slower than the former two steps, amine and isocyanate functionalities are not present simultaneously, preventing the formation of the symmetric urea. During this reaction, two equimolar amounts of both tert-butanol and carbon dioxide are released.
O O O O O
O O R NH2 + O O O O O O H + R N H
O-
CO2 t-BuOH
O R N
O O H
CO2 t-BuOH
R N C O
Scheme 1.2: Mechanism for the formation of isocyanates from di-tert-butyl tricarbonate and primary amines.
This synthetic procedure allowed the synthesis of unusual mono- and multiisocyanates with a minimum of side reactions. Some of these isocyanates were inaccessible by conventional methods, e.g. 1,3-propanediisocyanate, trans-1,2-diisocyanato-cyclohexane, and a dendrimer containing 64 terminal isocyanate groups. The only restraint of this reagent is the limited stability at room temperature of 1, especially in polar solvents, like methanol, pyridine or DMSO.
1.4 Spider dragline silk A famous thermoplastic elastomer that is found in Nature, is spider dragline silk.25 This high-performance protein fiber intrinsically combines both high tensile strength and high elasticity.26 The toughness (energy to break) of the spider silk exceeds the values for steel and polyaramids. In addition, the spider silk is produced under ambient conditions from aqueous solution in nature, while steel and polyaramids require a high temperature and a hazardous solvent, respectively, for processing. These properties have generated considerable interest in spider silk, especially in the dragline silk from Nephila clavipes, the golden orb weaver, whose silk is among the strongest (table 1.1). However, due to the small diameter of the fiber (max. 4 m), its strength on a macroscopic level is rather limited. Table 1.1: Tensile properties of some natural and synthetic fibers.25a Material Density Youngs Strength Strain Toughness modulus at break -3 (g/cm ) E (GPa) br (GPa) br (%) (kJ/kg) 1.1 1.4 0.97 7.8
a
Nylon 6,6 Kevlar 49 Dyneema Steel Spider silk
5 130 110 200 10
0.95 3.6 3.4 1.5 1.1
18 3 3.5 1 27
73 36 60 0.77 123
1.3
a) Dragline of Araneus diadematus. The outstanding properties of spider silk are a consequence of both the molecular architecture and the amino acid composition. The primary constituents of spider silk are the two simplest amino acids, glycine (42% of the structure) and alanine (25%). The remainder of the fiber consists of bulkier amino acids, with glutamine, serine, tyrosine, and leucine prominently represented.27 Sequencing of the protein, reveals that it consists of repeating units in which a polyalanine block of five to seven residues is followed by a glycine-rich sequence 7
Chapter 1
containing the bulky residues.28 From this point of view, it resembles the molecular structure of a segmented copolymer. The alanine block has been shown to adopt an antiparallel pleated -sheet conformation. After a detailed analysis by SAXS and solid-state 2H-NMR, Jelinski et al. proposed a structural model for the silk fiber, with highly-oriented alanine-rich crystals, and with weakly-oriented isolated -sheets.29 Both are dispersed in an elastic, non-ordered, glycine-rich phase (figure 5.1). Specific residues in the alanine-rich segments cause reversal of the chain direction, via a -hairpin. This inhibits the growth of the -sheet, resulting in the formation of very small crystals of 2 5 6 nm in size.29a Although the crystallinity of the material is quite low (30%), the abundance of these tiny reinforcing fillers accounts for the fibers toughness. This picture is consistent with a theoretical model of spider silk elasticity developed by Termonia.30
fiber axis
Figure 1.5: Schematic representation of a spider dragline silk fiber. Within the past few years, there has been a renewed interest in the use of spider dragline silk for commercial applications. Silk of the silkworm Bombyx mori is produced on large scale by cultivation of the silkworms, allowing the commercial use of this material as textile fiber. Unfortunately, spider silk cannot be produced in this manner due to cannibalism of the spiders. This challenged researchers to prepare the spider silk via (bio)synthetic methods. Laboratory-scale expression of silk proteins is now feasible, in both bacteria and yeasts.31 However, the most promising technique for large scale production is the expressing of silk genes in mammalian species.32 The company Nexia Biotechnologies is currently employing this recombinant DNA technique for the production of large amounts of spider silk (trade name Biosteel), by using transgenic goats that secrete the silk protein in their milk.33 Even though there is still a lot of debate about the morphology of spider dragline silk, it is generally accepted that its unique properties are the result of its well-defined morphology in combination with the spinning process from a liquid crystalline solution in the spiders 8
gland. This example of spider dragline silk offers insight into the behaviour of biomaterials, but more importantly, it teaches us, synthetic chemists, that precise control over the molecular structure and morphology, by applying well-defined building blocks and well-balanced secondary interactions, can give rise to novel materials possessing unique properties.
1.5 Well-defined thermoplastic elastomeric polymers The synthetic procedure to prepare TPUs, as described in section 1.2, has the intrinsic disadvantage that it leads to a distribution in the hard block lengths.34 In addition to this, the prepolymers that are generally used are polydisperse. Both facts result in a quite inhomogeneous chemical composition of the segmented copolymer. The dispersities in hard and soft block length directly influence the material properties of the polymers. The phase separation within these block copolymers is incomplete.1 Part of the hard blocks, in particular the shorter ones, are dissolved in the soft phase, causing an increase in the glass transition temperature, which is undesired for the low temperature flexibility of the material. The polydispersity of the hard block is manifested in a broad melting range and a rubbery plateau that is dependent on temperature. To improve the properties of TPEs, and to get more insight into the structure properties relationship, block copolymers containing hard blocks of uniform length have been prepared.35,36 Several types of uniform hard blocks were used, such as non-hydrogen bonding polyurethanes,35a,b normal polyurethanes,35c-h polyurethane ureas,35i and aramid units.36 In all these studies, the hard blocks were synthesized first. Due to their low solubility, especially of the longer ones, the maximum number of repeating units in the hard blocks is limited to approximately three repeating units. Subsequently, these hard blocks were polymerized with the appropriate soft blocks, to yield the segmented copolymers. The properties of these materials were compared to those of the analogous materials having polydisperse hard blocks, and these studies showed the advantages of the well-defined molecular structure. The materials showed higher crystallinity and better phase separation, causing the properties to be less dependent on temperature. The ultimate properties of the material improved considerably: higher strengths and strains at break were achieved. Some studies were preformed on the influence of the polydispersity of the soft block on the properties of thermoplastic elastomeric polyurethanes.35a,b,i,37 Harrell and Cooper described the synthesis and properties of block copoly(ether urethane)s, in which the urethane part was based on piperazine (figure 1.6). These secondary urethane groups are unable of forming hydrogen bonds. According to the authors, this should facilitate the interpretation of the structureproperties relationship. A pTHF prepolymer of low polydispersity (P.D.=1.1) 9
Chapter 1
was obtained, upon fractionation of a commercially available, polydisperse prepolymer. This was used as a monodisperse soft block, and built-in into the block copolymer. The material properties of this block copolymer were compared to those of an analogous block copolymer containing a polydisperse soft block (P.D.=1.7). The authors concluded that only some minor effects were observed upon lowering the polydispersity of the soft block: both strength and strain at break increased slightly.
O CN O O O NCO C4H8O y n NCO C4H8O CN x
Figure 1.6: Block copoly(ether urethane) possessing secondary urethane groups. Shirasaka et al., who studied block copoly(ether urethane urea)s possessing monodisperse soft blocks (P.D.1.1) based on pTHF, found comparable results.35i Although this polymer showed a slightly lower strength and strain at break, the effects observed upon lowering the polydispersity of the soft block to approximately 1.1 were not significant. At this point the question arises, whether it would make sense to use soft blocks with even lower polydispersity, in other words, truly monodisperse soft blocks. Some examples of truly monodisperse polymers or block copolymers are described in literature.38,39 Tirrell et al. studied the liquid crystalline ordering of monodisperse poly(-benzyl ,L-glutamate), that was prepared via protein engineering.38 Polydisperse samples of this rod-like polymer order in a nematic phase, whereas the monodisperse analogue shows the quite rare smectic ordering. Triblock copolymers possessing a monodisperse rod-like block were prepared by Stupp et al.39 Their self-assembly into discrete mushroom-shaped nanosized objects is a direct result of the well-defined character of the triblock copolymer. In conclusion, the examples of Tirrell and Stupp show that truly monodisperse, well-defined polymers may give rise to highly ordered microstructures, which is even manifested on a macroscopic scale. Two examples are reported of segmented block copolymers possessing a uniform distribution in both hard and soft segment length.37 Such block copoly(ether ester)s were prepared by Wegner et al. (figure 1.7).37a,b They synthesized uniform butylene terephthalateoligomers, as well as monodisperse pTHF-oligomers, via a stepwise synthesis. Both building blocks were coupled via a solution polycondensation to avoid transesterification, which would undo the uniformity of the hard blocks. The melting behaviour of the polymers was studied in detail. However, the limited amounts available of these polymers, due to the tedious purifications of the building blocks, hampered the study of their mechanical properties.
10
O C
O CO C4H8O y n
CO C4H8O C x
Figure 1.7: Block copoly(ether ester)s with uniform segments; x=14, y=12. Eisenbach and Baumgartner prepared block copoly(ether urethane)s comprising both uniform hard and soft segments (figure 1.8),37c using the same procedure as Wegner et al. The uniform block copoly(ether urethane)s were studied in detail with special emphasis on the thermal behaviour and hydrogen bonding. Again, no mechanical properties were discussed, unfortunately. They concluded that these well-defined block copolymers show better phase separation, and hence, sharper phase transitions.
O CN H CH2 O O CH2 H O NCO C4H8O y n NCO C4H8O CN x H H
Figure 1.8: Block copoly(ether urethane)s with uniform segments; x=14, y=12.
1.6 Aim of this thesis This introduction gave a brief overview of thermoplastic elastomers, and thermoplastic elastomeric polyurethanes more specifically. The synthetic procedure employed to prepare these materials, results in polymers possessing an inhomogeneous microstructure: both the hard and soft blocks are polydisperse. These features have a negative effect on the material properties. On the other hand, spider dragline silk demonstrates us how exact control over molecular architecture and secondary interactions can give rise to materials with very appealing properties. The aim of this thesis is to prepare well-defined thermoplastic elastomers from uniform and monodisperse building blocks, by implying precisely-controlled supramolecular interactions (hydrogen bonding, electrostatic interactions). By studying their morphology and material properties in detail, more insight should be gained in the structureproperties relationship of these materials. Employing the novel isocyanate chemistry and highlyselective syntheses, we wish to prepare well-defined building blocks for these polymers. In this way, we expect to close the gap between ill-defined synthetic thermoplastic elastomers and well-defined biomaterials.
11
Chapter 1
1.7 Outline of this thesis The potential of the novel isocyanate chemistry, employing di-tert-butyl tricarbonate, in the field of polymer chemistry is described in chapter 2. Starting from -amino--alcohols, the (so far) unknown class of [n]-polyurethanes is developed and successfully characterized. Some properties of these polymers were studied. In addition to that, hyperbranched polyurethanes were synthesized and characterized in detail. This isocyanate chemistry is utilized in the next chapters for the synthesis of thermoplastic elastomers. In chapter 3, the synthesis and modification of pTHF-based prepolymers is described. These were used as the soft segment in block copoly(ether urea)s, possessing uniform hard blocks comprising 1 4 urea groups. A detailed molecular characterization of these polymers was performed, which showed the successful crosslinking by hydrogen bonding between hard blocks comprising two or more urea groups. The block copoly(ether urea)s possess thermoplastic elastomeric properties. The thermal characteristics, the morphology, and the mechanical properties were studied by a variety of experimental techniques in chapter 4. The behaviour of the materials upon deformation was examined and related to the structural changes on the molecular level. Due to the uniformity of the bisureido-butylene hard block, it was possible to anchor molecules containing the complementary bisureido-butylene unit to these polymers. In chapter 5, a modular approach towards polymer modification based on this principle is described by means of two examples. In chapter 6, monodisperse pTHF-oligomers are described that were synthesized via a stepwise procedure on a multigram scale, resulting in a pTHF 13-mer (MW=954 g/mol) of very low polydispersity (P.D.=1.004). This oligomer was used for the preparation of a block copoly(ether urea) possessing a uniform hard and a monodisperse soft block. Its properties are compared to those of its less-defined analogue. Finally, in chapter 7, a novel type of reversible networks is described, based on the association of the end groups of a telechelic polymer and the host groups of a multivalent dendrimer. The solution properties of this system are studied in detail by viscometry and dynamic light scattering. Upon increasing the concentration of the complex, a transition from flower-like structures to a transient network is observed. The influence of the molecular structure of the guest on this transition is investigated.
12
1.8 References and notes

1. Legge, N. R., Holden, G. and Schroeder, H. E. Thermoplastic elastomers: A comprehensive review; Carl Hansser Verslag: New York, 1987. 2. Mark, H. F. Encyclopedia of polymer science and technology; 3 ed.; Wiley-Interscience: New York, 2001. 3. a) Kennedy, J. P. and Castner, K. F. J. Pol. Sci. A 1979, 17, 2055; b) Engle, L. P. and Wagener, K. B. J. Macromol. Sci. Rev. 1993, C33, 239. 4. Ullmann Ullmann's Encyclopedia of industrial chemistry; 6 ed.; VCH: Weinheim, 2001. 5. Niesten, M. Ph.D. Thesis, University of Twente (Enschede), 2000. 6. Holden, G., Bishop, E. T. and Legge, N. R. J. Pol. Sci. C 1969, 26, 36. 7. a) Coleman, D. J. Polym. Sci. 1954, 14, 15; b) Witsiepe, W. K. ACS Advances in Chemistry 1973, 129, 39. 8. Deleens, G., Foy, P. and Marechal, E. Eur. Pol. Journal 1977, 13, 337. 9. Bayer, O., Muller, E., Petersen, S., Piepenbrink, H. F. and Windemuth, E. Angew. Chem. 1950, 62, 57. 10. Cella, R. J. J. Polym. Sci.: Symp. 1973, 42, 727. 11. a) Wirpsza, Z. Polyurethanes: Chemistry, Technology and Applications; Ellis Horwood: London, 1993; b) Saunders, J. H. and Frisch, K. C. Polyurethanes : chemistry and technology, part 1 Chemistry; Interscience: New York, 1962; Vol. 16; c) Saunders, J. H. and Frisch, K. C. Polyurethanes : chemistry and technology, part 2 Technology; Interscience: New York, 1964; Vol. 16; d) David, D. J. and Stanley, H. B. Analytical chemistry of polyurethanes, part 3; Wiley-Interscience: New York, 1969; Vol. 16. 12. a) Spaans, C. J. Ph.D. Thesis, University of Groningen (Groningen), 2000; b) Runt, J., Xu, R., Manias, E. and Snyder, A. J. Macromolecules 2001, 34, 337. 13 Ulrich, H. Chemistry and technology of isocyanates; J. Wiley & Sons: Chichester, 1996. 14. a) Etter, M. C., Urbanczyk-Lipkowska, Z., Zia-Ebrahimi, M. and Panunto, T. W. J. Am. Chem. Soc. 1990, 112, 8415; b) Desiraju, G. R. Organic solid state chemistry; Elsevier: Amsterdam, 1987; Vol. 32; c) Carr, A. J., Melendez, R., Geib, S. J. and Hamilton, A. D. Tetrahedron Lett. 1998, 39, 7447; d) Pathirana, H. M. K. K., Weiss, T. J., Reibenspies, J. H., Zingaro, R. A. and Meyers, E. A. Z. Kristallogr. 1994, 209, 696; e) Perez-Folch, J., Subirana, J. A. and Aymami, J. J. Chem. Cryst. 1997, 27, 367. 15. Bayer, O. Angew. Chem. 1947, 59, 257. 16. Bachmann, F., Reimer, J., Ruppenstein, M. and Thiem, J. Macromol. Rapid Commun. 1998, 19, 21. 17. a) Carothers, W. H. (DuPont) Patent US2130948, 1938; b) Carothers, W. H. (DuPont) Patent US2071253, 1937. 18. Witsiepe, W. L. (DuPont) Patent US3377322, 1968. 19. Bhat, G., Chand, S. and Yakopson, S. Thermochim. Acta 2001, 367-368, 161. 20. Reisch, M. Chem. Eng. News 1999, 77, 70. 21. a) Siefken, W. Justus Liebigs Ann. Chem. 1949, 562, 75; b) Ozaki, S. Chem. Rev. 1972, 72, 457. 22. Waldman, T. E. and McGhee, W. D. Chem. Comm. 1994, 8, 957. 23. Peerlings, H. W. I. and Meijer, E. W. Tetrahedron Lett. 1999, 40, 1021. 24. Pope, B. M., Yamamoto, Y. and Tarbell, D. S. Org. Synth. 1978, 57, 45. 25. a) Kubik, S. Angew. Chem. Int. Ed. 2002, 41, 2721; b) Tirrell, D. A. Science 1996, 271, 39. 26. a) Shao, Z. and Vollrath, F. Polymer 1998, 40, 1799; b) Madsen, B., Shao, Z. Z. and Vollrath, F. Int. J. Biol. Macromol. 1999, 24, 301; c) Gosline, J. M., Denny, M. W. and DeMont, M. E. Nature 1984, 309, 551; d) Gosline, J. M., Shadwick, R. E., Demont, M. E. and Denny, M. W. Biol. Synth. Polym. Networks 1988, 57; e) Gosline, J. M., Guerette, P. A. and Ortlepp, C. S. J. Exp. Biol. 1999, 202, 3295; f) Cunniff, P. M., Fossey, S. A., Auerbach, M. A., Song, J. W., Kaplan, D. L., Adams, W. W., Eby, R. K., Mahoney, D. and Vezie, D. L. Polymers for Advanced Technologies 1994, 5, 401. 27. Fukushima, Y. Biopolymers 1998, 45, 269. 28. Xu, M. and Lewis, R. V. Proc. Natl. Acad. Sci. USA 1990, 87, 7120. 29. a) Yang, Z., Grubb, D. T. and Jelinski, L. W. Macromolecules 1997, 30, 8254; b) Simmons, A. H., Michal, C. A. and Jelinski, L. W. Science 1996, 271, 84. 30. a) Termonia, Y. Macromolecules 1994, 27, 7378; b) Termonia, Y. Pergamon Materials Series 2000, 4, 269.
13
Chapter 1
31. a) Fahnestock, S. R. and Irwin, S. L. Appl. Microbiol. Biotechnol. 1997, 47, 23; b) Fahnestock, S. R. and Bedzyk, L. A. Appl. Microbiol. Biotechnol. 1997, 47, 33; c) O'Brien, P., Fahnestock, S. R., Termonia, Y. and Gardner, K. H. Adv. Mater. 1998, 10, 1185; d) Prince, J. T., McGrath, K. P., DiGirolamo, C. M. and Kaplan, D. L. Biochemistry 1995, 34, 10879; e) Arcidiacono, S., Mello, C., Kaplan, D., Cheley, S. and Bayley, H. Appl. Microbiol. Biotechnol. 1998, 49, 31. 32. Cohen, P. New Sci. 1998, 160, 11. 33. For more information see: www.nexiabiotech.com . 34. Sorta, E. and Melis, A. Polymer 1978, 19, 1153. 35. a) Harrell, L. L. Macromolecules 1969, 2, 607; b) Ng, H. N., Allegrezza, A. E., Seymour, R. W. and Cooper, S. L. Polymer 1973, 14, 255; c) Eisenbach, C. D. and Stadler, E. Macromol. Chem. Phys. 1995, 196, 1981; d) Miller, J. A., Shaow, B. L., Hwang, K. K. S., Wu, K. S., Gibson, P. E. and Cooper, S. L. Macromolecules 1985, 18, 32; e) Fu, B., MacKnight, W. J. and Schneider, N. S. Rubber Chem. Technol. 1986, 59, 896; f) Qin, Z. Y., Macosko, C. W. and Wellinghoff, S. T. Macromolecules 1985, 18, 553; g) Furukawa, M., Komiya, M. and Yokoyama, T. Angew. Makromol. Chem. 1996, 240, 205; g) Lai, Y.-C., Quinn, E. T. and Valint, P. L., Jr. J. Pol. Sci. A 1995, 33, 1767; h) Blundell, D. J., Eeckhaut, G., Fuller, W., Mahendrasingam, A. and Martin, C. Polymer 2002, 43, 5197; i) Shirasaka, H., Inoue, S.-i., Asai, K. and Okamoto, H. Macromolecules 2000, 33, 2776. 36. a) Gaymans, R. J. and Haan, J. L. d. Polymer 1993, 34, 4360; b) Niesten, M. C. E. J., Gaymans, R. J. and Brinke, A. t. Polym. Prepr. 1999, 40, 1012; c) Niesten, M. C. E. J., Gaymans, R. J. and Brinke, A. t. Polym. Prepr. 1999, 40, 1012; d) Niesten, M. C. E. J., Harkema, S., van der Heide, E. and Gaymans, R. J. Polymer 2000, 42, 1131; e) Niesten, M. C. E. J., Tol, R. and Gaymans, R. J. Polymer 2000, 42, 931; f) Niesten, M. C. E. J., Feijen, J. and Gaymans, R. J. Polymer 2000, 41, 8487; g) Niesten, M. C. E. J. and Gaymans, R. J. J. Appl. Polym. Sci. 2001, 81, 1372; i) Niesten, M. C. E. J. and Gaymans, R. J. Polymer 2001, 42, 6199. 37. a) Schmidt, H. G. Ph.D. Thesis, University of Freiburg (Freiburg), 1984; b) Wegner, G., in Legge, N. R., Holden, G. and Schroeder, H. E. Thermoplastic elastomers: A comprehensive review; Carl Hansser Verslag: New York, 1987, Section 12/5; c) Eisenbach, C. D., Baumgartner, M. and Gunter, G. Adv. Elastomers Rubber Elasticity, [Proc. Symp.] 1985, 51. 38. a) Yu, S. M., Conticello, V. P., Zhang, G., Kayser, C., Fournier, M. J., Mason, T. L. and Tirrell, D. A. Nature 1997, 389, 167; b) Yu, S. M., Soto, C. M. and Tirrell, D. A. J. Am. Chem. Soc. 2000, 122, 6552. 39. a) Radzilowski, L. H., Wu, J. L. and Stupp, S. I. Macromolecules 1993, 26, 879; b) Radzilowski, L. H. and Stupp, S. I. Macromolecules 1994, 27, 7747; c) Radzilowski, L. H., Carragher, B. O. and Stupp, S. I. Macromolecules 1997, 30, 2110, d) Stupp, S. I., LeBonheur, V., Walker, K., Li, L. S., Huggins, K. E., Keser, M. and Amstutz, A. Science 1997, 276, 384; e) Zubarev, E. R., Pralle, M. U., Li, L. and Stupp, S. I. Science 1999, 283, 523.
14
2
[n]-Polyurethanes and hyperbranched polyurethanes
Abstract A facile and mild synthetic procedure for the preparation of isocyanates, using di-tert-butyl tricarbonate, is employed to synthesize -isocyanato--alcohols 3 from -amino--alcohols 2. These isocyanato alcohols were polymerized in situ to yield the aliphatic [n]-polyurethanes, polymers that were still unknown as a general class, despite its structural simplicity. The [n]-polyurethanes are highly crystalline, brittle polymers, and show a strong alternation in their melting points with respect to the number of methylene groups in the monomer unit. The polymers with an even number of methylene units, possess a higher melting point than those with an odd number. Using the same synthetic procedure, also aliphatic hyperbranched polyurethanes were prepared, starting from AB2 and A2B monomers. These polymers were obtained in high yields as viscous oils. A detailed 13C-NMR analysis of one of the hyperbranched polyurethanes indicated that its degree of branching is quite low, DB=0.31 0.06. This isocyanate chemistry is utilized in the next chapters for the synthesis of thermoplastic elastomers.
15
Chapter 2
2.1 Introduction Polyurethanes form a versatile class of polymers, which are used in a broad range of applications1, like automotive, foams, coatings, fibers, and biomedical applications.2 The properties of polyurethanes depend strongly on their macromolecular structure, i.e. the nature and functionality of its constituting (macro)monomers. Covalently cross-linked networks are thermosets, and are used as both rigid and flexible foams; segmented polyurethanes show thermoplastic elastomeric properties, and are used e.g. as textile fibers or in hoses; linear homopolyurethanes are thermoplastic materials, and can be used as fibers3 or as biodegradable material.4
O HO CH2
m
O N H CH2
n-2
OH +
OCN
CH2
n-2
NCO
N H
CH2
Scheme 2.1: Synthesis of linear [m,n]-polyurethanes. In most, if not all, cases of linear polyurethanes, the macromolecular structure is based on the reaction of dihydroxy-compounds with diisocyanates, yielding A2B2-polymers or the [m,n]-polyurethanes (scheme 2.1). The values of m and n represent the total number of carbon atoms in the monomeric unit of the dihydroxy-compound and the diisocyanate, respectively. A comprehensive study of the parent aliphatic [m,n]-polyurethanes was reported by Otto Bayer in 1947.3
H N O
O N H O H N O
H N O
H N O
N H O
H N O O
H N O O
O O N H
H N O O
H N O O
N H O
[m,n]-nylon
[n]-nylon
[m,n]-polyurethane
[n]-polyurethane
Scheme 2.2
16
These, at that time, novel structures were compared and proposed to compete with the two series of aliphatic polyamides, the [n]- and [m,n]-nylons (scheme 2,2), which were synthesized by Wallace Carothers in the 1930s.5 The [n]-nylons (AB-polyamides) are either synthesized from -amino--carboxylic acids or from cyclic amides; the [m,n]-nylons (A2B2polyamides) are prepared from diamines and dicarboxylic acids. Whereas the disclosure of the [n]-nylons quickly followed Carothers first description of the [m,n]-nylons, the general class of linear aliphatic [n]-polyurethanes is notably absent in the impressive list of linear macromolecules, despite its structural simplicity and strong resemblance to the [n]-nylons. Why is it, that these aliphatic [n]-polyurethanes have not been prepared before, despite their simple structure? Obviously, this is due to the unavailability of the appropriate monomers; both the -isocyanato--alcohols and the cyclic carbamates are not known and synthesized as a general class. As a result the aliphatic series is limited to a few specific structures.6 Only recently, and after our work was published, Hcker et al. described the synthesis of some [n]-polyurethanes, with n=37 using activated -amino--alcohols7, or ring opening polymerization of trimethylene urethane.8 In the past, some oligomers have been synthesized via a stepwise sequence.9 The aromatic poly(1,4-phenylene urethane) has been prepared by polymerization of 4-isocyanato-phenol that was obtained via the Curtius rearrangement of 4-hydroxybenzoyl azide, as was reported by Kinstle and Sepulveda, and by several others.10 However, the much higher nucleophilicity of aliphatic alcohols over phenols requires a much milder conversion of the amino group of -amino--alcohols into an isocyanato--alcohols, followed by a controlled polymerization. Unfortunately, most of the methods known to transform amines into isocyanates e.g. with phosgene or other suitable carbonic acid derivatives are not mild enough and will furnish several side products containing carbonate or urea groups which are the result of uncontrolled reactions of the two nucleophilic species being able to react with the carbonic acid derivative. Peerlings reported on the potential of di-tert-butyl tricarbonate 1, as a versatile and mild reagent for the synthesis of isocyanates.11 Di-tert-butyl tricarbonate 1 is synthesized from carbon dioxide, potassium tert-butoxide, and phosgene,12 and is readily obtained in multigram quantities. Upon reaction of aliphatic or some aromatic amines with a stoichiometric amount of 1, a fast reaction occurs, in which the amine functionality is converted to an isocyanate within a few minutes at room temperature (scheme 2.3), see section 1.3 for more details. During this reaction, two equimolar amounts of both tert-butanol and carbon dioxide are released. This synthetic procedure allowed the synthesis of unusual mono- and multi-isocyanates with a minimum of side reactions. Some of these isocyanates were inaccessible by conventional techniques, e.g. 1,3-propanediisocyanate, trans-1,2diisocyanato-cyclohexane, and a dendrimer containing 64 terminal isocyanate groups. The
17
Chapter 2
only restraint of this reagent is the limited stability at room temperature, especially in polar solvents, like methanol, pyridine or DMSO.
O R NH2 + O O O O O O R N C O + 2 CO2 + 2 t-BuOH
Scheme 2.3: Synthesis of isocyanates from di-tert-butyl tricarbonate and primary amines. This novel procedure to prepare isocyanates under very mild conditions prompted us to use it for the synthesis of -isocyanato--alcohols, the monomer for the [n]-polyurethanes, starting from -amino--alcohols. In this chapter, a general and convenient route to aliphatic -isocyanato--alcohols and their in situ polymerization into the corresponding [n]-polyurethanes are described.
2.2 Synthesis of -amino--alcohols -Amino--alcohols were used as the starting material for the synthesis of isocyanato--alcohols, the monomer for the [n]-polyurethanes. Short amino alcohols with x = 2 6 are commercially available, in contrast to their longer homologs, which had to be prepared.
HO (CH 2) NH 2
x
2-(x)
Amino alcohols with x = 9, 10, and 11 were prepared via the Gabriel synthesis13 by the hydrazine deprotection of -phthalimide--alcohols, which were made out of the commercially available -bromo--alcohols (scheme 2.4).
O HO (CH 2) Br + K N x O
+ _
KBr CH 3CN
x
O HO (CH 2) N O N 2H 4 EtOH HO (CH 2 ) NH 2

x
Scheme 2.4: Synthesis of amino alcohols via the Gabriel synthesis, x = 9, 10, and 11. This synthesis is rather straightforward, and after distillation, the amino alcohols were obtained in moderate yield (60 80%). The products were successfully characterized by 1H-, 13 C-NMR, GC, elemental analysis and their melting points. Amino alcohols with x = 7, 8, and 12 were prepared from the corresponding lactams (scheme 2.5). Acidic hydrolysis of the lactam resulted in the open amino carboxylic acids, 18
which were derivatized to their ethyl esters. Both reactions proceeded in almost quantitative yield. The last step is a reduction of the ester to an alcohol using lithium aluminum hydride. Due to the basic conditions of this reduction, a considerable amount of secondary amide was formed, resulting in a mixture of the product, the starting lactam, and several oligoamides. Distillation of the crude products yielded the amino alcohols in a rather poor yield (30 40%). The products were successfully characterized using the same techniques as described above.
O H C N
x-1
(CH2)
HCl 100C
HOOC (CH 2) NH 3Cl

x-1
SOCl2 EtOH
EtOOC (CH2) NH3 Cl

x-1
LiAlH 4 HO (CH2 ) NH 2 x THF 2
Scheme 2.5: Synthesis of amino alcohols derived from lactams, x = 7, 8, and 12.
2.3 Preparation of [n]-polyurethanes The high selectivity and reactivity of di-tert-butyl tricarbonate 1 was used to synthesize -isocyanato--alcohols 3 from the prepared -amino--alcohols 2 (scheme 2.6).
O HO (CH 2 ) NH 2
x
O O 1 O
O O
CO 2, t-BuOH CHCl3, 20C
HO (CH 2 ) NCO
x
Scheme 2.6: Synthesis of -isocyanato--alcohols. This reagent is the key element for the selective formation of the intermediate isocyanato alcohols 3. Reaction of a small excess of 1 with 2 in chloroform at room temperature gave the isocyanato alcohol monomer 3 within 10 minutes. This reaction is accompanied by the formation of two moles of both carbon dioxide and tert-butanol. The former escapes from the solution, while the latter remains in the reaction mixture. Under the conditions employed here, tert-butanol is unreactive to the isocyanate, hence, it is harmless. The amino alcohol-solution was injected under the surface of a stirred solution of 1.05 equivalents of di-tert-butyl tricarbonate to avoid contact of the amino alcohol with carbon dioxide, since this may result in the formation of a carbamic acid which prevents further reaction. This side-reaction is notified by a turbidity in the reaction mixture. After decomposition of the unstable carbamic acid into the initial amino alcohol, the latter reacts with already formed isocyanate, resulting in symmetrical urea. This side reaction distorts the perfect stoichiometry of the AB-type polymerization, and consequently, limits the molecular weight of the polymer. For the short amino alcohols with x = 2 or 3, the synthesis of the respective isocyanato alcohols failed, due to their spontaneous ring closure into cyclic urethanes with five,
19
Chapter 2
respectively six ring-atoms. For the longer amino alcohols, the formation of the isocyanato alcohols was confirmed by infrared and 1H-NMR spectroscopy. In the IR spectrum, a strong absorption corresponding to the N=C=O stretch-vibration at 2274 cm1 was observed for solutions of 3 in chloroform. Figure 2.1b shows the 1H-NMR spectrum of 5-isocyanato-1pentanol after reaction of 5-amino-1-pentanol (Figure 2.1a) with di-tert-butyl tricarbonate.
a)
(ppm)
t-BuOH
t-BuOH
b)
H2O
DMSO-d6 urea
c)
Figure 2.1: 1H-NMR spectra of a) 5-amino-1-pentanol, b) 5-isocyanato-1-pentanol, c) [6]polyurethane. Figure 2.1b reveals the absence of any side products, and also proves the relative stability of 3 in solution. However, evaporation to dryness furnished undefined products. The polymerization of monomers 3 is performed in situ by addition of a catalytic amount of zirconium(IV) acetylacetonate or dibutyltin dilaurate (scheme 2.7).
O HO (CH2) NCO
x Zr(acac)4 CHCl3, 20C
O O
n
(CH 2) N x H 4-(x) x = 4 12
or
HN
(CH 2)x x=23
Scheme 2.7: Polymerization of isocyanato alcohols to [n]-polyurethanes. 20
The precipitation of polymers 4 was observed within a few hours. The polymers were isolated by filtration in a yield of about 60% as a white microcrystalline powder. Since the polymerization was performed in solution, a relatively large amount of soluble, cyclic oligomers was formed. ElectroSpray Mass-Spectrometry of the filtrate showed the presence of cyclic dimers up to cyclic hexamers. According to NMR, FT-IR and elemental analyses, all polymers possess a very uniform microstructure and these techniques confirm the assigned structures. With 1H-NMR spectroscopy (figure 2.1c) both the syn and anti carbamate conformations14 are observed. Further, it revealed that the polymers contain a very small amount (less than 2%) of urea linkages, probably formed by hydrolysis of the isocyanate functionality during reaction. The molecular weights of the polymers were determined by both size exclusion chromatography (SEC) and Ubbelohde viscometry. The SEC measurements were performed in N-methyl pyrrolidone (NMP) as solvent, and polystyrene was used as reference. The intrinsic viscosities of the polymers were determined in m-cresol, and the MarkHouwink parameters of the known [m,n]-polyurethanes were used to estimate the molecular weights of these comparable [n]-polyurethanes.1 The data obtained with SEC corresponds nicely with the values found with viscometry (table 2.1). Table 2.1 Characteristics of [n]-polyurethanes 4. x 4 4 5 5 5 5 6 7 8 9 10 11 12 5-co-6 Solvent CHCl3 DMSO CHCl3 CHCl3 THF DMSO CHCl3 CHCl3 CHCl3 CHCl3 CHCl3 CHCl3 CHCl3 CHCl3 T (C) 20 20 20 61 20 20 20 20 20 20 20 20 20 20 Yield (%) 61 63 70 43 65 56 55 61 57 63 59 48 Mw (kg mol1)a 18.9 12.5 22.3 29.5 15.3 10.6 34.4 26.5 30.5 67.4 32.6 20.3 21.4 22.7 Mw/Mna 1.4 1.7 1.5 1.5 1.4 1.3 1.4 1.5 1.6 1.6 1.7 1.5 1.6 1.4 [] (dl g1) 0.16 0.17 Mv (kg mol1)b 16 18
0.22 0.24 0.24 0.43 0.30 0.18 0.26
25 29 29 64 39 19 32
a) Measured by SEC, with NMP as solvent, relative to polystyrene standards. b) Calculated from intrinsic viscosity in m-cresol at 25C; K=1.34*104, a=0.73.
21
Chapter 2
The weight-averaged molecular weights of the polymers, varying between 20 and 70 kg mol , are fairly high to respectable, despite the precipitation of the polymers from the poor solvent during preparation. Performing the polymerization in more polar solvents like THF, or
-1
DMSO, or at reflux temperature in chloroform, does not significantly change the molecular weight of the polymers, as investigated. Table 2.1 shows the yields of the [n]-polyurethanes and estimations of their molecular weights. 2.4 Thermal properties of [n]-polyurethanes The thermal properties of the polymers were studied by DSC, indicating that all polymers are highly crystalline. Figure 2.2 displays the second heating and cooling curve of [6]-polyurethane at 10C/min. The thermogram shows a typical curve reminiscent of a semicrystalline polymer with a broad melting peak, and a shoulder at lower temperatures indicative of a recrystallization process prior to melting. The relative value of this recrystallization peak can be altered by changing the cooling rate or by annealing the sample. A small undercooling is observed of about 20-30C for all [n]-polyurethanes, indicating that these polymers have a high rate of crystallization.
22 21.5 Heat Flow Endo Up (mW) 21 20.5 20 19.5 19 50 70 90 110 T (C) 130 150 170 190 Tc=100.1C H=38.2 J/g Tm=127.3C H=41.6 J/g
Figure 2.2: DSC curve of [6]-polyurethane. After melting and crystallization of the microcrystalline powders, as obtained after the polymerization, the polymer is obtained as a pale brittle material. Melting points and melting enthalpies of the polymers are depicted in table 2.2, and melting points are plotted in figure 2.3. The melting points are much lower than those of the corresponding [n]-nylons, as is also observed for [m,n]-polyurethanes compared to [m,n]nylons. It is evident from figure 2.3, that the [n]-polyurethanes show a strong odd-even effect in their melting points. In conformity with other polymers, e.g. [n]-nylons,15 the [n]22
polyurethanes with an odd number of methylene units in the monomeric main chain melt at lower temperatures than the [n]-polyurethanes with an even number of methylene units. At first glance it is surprising that it is not the total number of main chain atoms between the hydrogen bonding units that determines the odd-even effect. However, many examples are known in which the number of methylene units determines this effect.16 The alternation of the melting points is caused by the packing of the alkyl spacer in the crystalline phase. Boese et al. studied this behaviour in detail for several homologous series of substituted alkanes.16b-e Their conclusion is that compounds with an even number of methylene units in the alkyl spacers pack more densely than the odd analogues. This results is a higher melting point for the compounds with an even number of methylene units.
200 180 Tm(C) 160 140 120 100
8 x
10
12
Figure 2.3: Melting points of [n]-polyurethanes: (n) experimental, (g) calculated; x denotes the number of methylene units in the monomer unit. The observed melting points are 10 to 20 degrees lower than those predicted by the additive group theory developed by van Krevelen.17 It is noteworthy to mention that the melting points can be raised by approximately 5C when the low molecular weight fraction is removed by extensive Soxhlet extraction with hot methanol. Table 2.2 DSC measurements of [n]-polyurethanes 4; heating rate of 10C min1. x Tm(C) Hm(J/g) 4 194 89.7 5 127 41.6 6 157 102.8 7 114 51.3 8 146 78.8 9 116 56.7 10 148 65.7 11 119 61.9 12 141 68.5
A random copolymer derived from 5-isocyanato-pentanol and 6-isocyanato-hexanol in a 1:1 molar ratio was prepared in the same way as the homopolymers. In contrast to the
23
Chapter 2
homopolymers, the random copolymer is soluble in chloroform, and is completely amorphous with a glass transition temperature of 8C according to DSC. According to thermogravimetric analysis (TGA), all polymers are stable up to 200C (figure 2.4), above which they decompose due to the cleavage of the urethane entities. The small decrease in weight slightly above 100C is believed to be due to loss of water.
60
100
140
180
220
260
300
T (C)
Figure 2.4: TGA of [6]-polyurethane.
2.5 Synthesis of hyperbranched polyurethanes Highly branched polymeric structures have received much attention from both chemists and physicists during the last decade. These structures combine appealing properties like a high degree of functionality, low melt-viscosity, and excellent solubility, thus making them interesting for a broad range of applications.18 In general, two distinct synthetic strategies have been employed for the preparation of such structures. One is a stepwise organic approach in which the macromolecules are built-up layer by layer, either in a convergent or in a divergent approach, resulting in monodisperse polymers with unprecedented structural precision.19 However, the preparation of these macromolecules, so called dendrimers, implies a multi-step synthesis, which hampers their practical use. The other synthetic strategy concerns the self-condensation of a ABx type monomer, with x2, resulting in a so called hyperbranched polymer.20 This polymeric approach has the advantage that it is synthetically more accessible and easy to scale-up in a one-pot synthesis. In contrast to the former approach, this latter approach yields macromolecules with less welldefined structures, having a polydisperse molecular weight distribution, and imperfect branching points. The properties of hyperbranched polymers tend to be intermediate between those of the fully regular dendrimers and those of linear polymers. 24
Due to the random and unpredictable nature of the branching reaction and the polydispersity inherent in all polymerizations, hyperbranched polymers contain a broad variety of molecular architectures. The molecular weight distribution is only one parameter to describe such a complex mixture of structural diversities.21 More architectural information is obtained by determining the degree of branching (DB) within the mixture. For AB2 monomers, this parameter was originally defined by Frchet (equation. 2.1)22: T +D (eq. 2.1) DB = T +D+L In eq. 2.1, T is the number of subunits in the terminal position (both end groups have not reacted), D is the number in a dendritic position (both end groups have reacted), and L is the number in a linear position (only one of the end groups has reacted). Thus, for a linear polymer, DB=0; and for a perfect dendrimer, DB=1. For a hyperbranched polymer, DB is in between both extremes, and the higher it is, the more branched and globular its structure is. In case of an ideal random polymerization of AB2 monomers, in which the reactivities of both Bgroups are equal and independent of one another, DB=0.5. If the reactivity of the second Bgroups is higher than the first one, more dendritic units are formed, and DB will increase. Contrary, a lower reactivity of the second B-group, results in an increase of the number of linear units, hence DB will decrease, and the polymer is less hyperbranched. For complete conversion of functional group A in the AB2 monomer, T and D are equal, and the more general equation 2.2 is obtained:23 L L 2D (eq 2.2) DB = = 1 = 1 N0 2D + L 2D + L were N0=T+D+L, the total number of monomer units. Thus, the degree of branching can be determined if at least one type of branching unit can be quantified. This can be done by using spectroscopic,20a, 25b or chromatographic techniques.24 A large variety of hyperbranched polymers has been reported, namely polyesters, polyureas, polyamides, polyphenylenes, polysiloxanes, polycarbonates, and polyethers.20a,b,25 Although, hyperbranched polyurethanes have been prepared,26 they are all containing aromatic structures, obtained via the Curtius rearrangement of benzoyl azides, or the dissociation of blocked isocyanates. The incorporation of the rigid aromatic rings in the hyperbranched polymers results in an increase of the glass transition temperature to far above room temperature. Fully aliphatic hyperbranched polyurethanes would have a low glass transition temperature, which makes them more easily processable. The synthetic procedure for the preparation of [n]-polyurethanes was also employed for the synthesis of aliphatic hyperbranched polyurethanes by using AB2 (5) and A2B (6) monomers. The monomers were synthesized by performing a Michael addition of acrylonitrile (ACN) to diethanolamine and 3-amino-propanol, respectively (scheme 2.8).
25
Chapter 2
After catalytic hydrogenation of the nitrile groups, aminodiol 5 and diamino alcohol 6 were obtained.
OH HN OH
ACN
OH NC N OH
H2, 80 bar Pd/C
OH H 2N 5 N OH
CN HO NH2
2 ACN
NH2
H2, 80 bar Pd/C
HO
N CN
HO
N 6 NH2
Scheme 2.8: Synthesis of monomers for hyperbranched polymers. Reaction of 5 and 6 with di-tert-butyl tricarbonate gave the corresponding isocyanate monomers, and in situ polymerization by adding Zr(acac)4 yields the hyperbranched structures 7, and 8 respectively (scheme 2.9). To prevent side reactions and crosslinking of the isocyanate end groups of hyperbranched polymer 8, they were derivatized by reaction with methanol.
OH 5 tricarb. CHCl3 OCN N OH Zr(acac)4 OCN ( N 7 O OH ) n-x O N )n H O NCO 6 tricarb. CHCl3 HO N NCO Zr(acac)4 2) MeOH HO ( N 8 N H N H O O ) CH3 n OCH3 ) n-x
Scheme 2.9: Synthesis of aliphatic hyperbranched polyurethanes 7 and 8. The hyperbranched polyurethanes are very viscous liquids, that are well soluble in i.e. chloroform, methanol, DMSO, and acidic water. Although SEC is not the appropriate technique to measure molecular weights of hyperbranched and dendritic structures, since their conformations are very different from the linear polystyrene, it was used to obtain a rough estimation of the molecular weights. According to SEC in NMP, the weight average 26
molecular weights of 7 and 8, are 6.4*103, and 3.8*103 g/mol, respectively, relative to polystyrene standards. The hyperbranched polyurethanes were characterized by 1H-NMR, 13 C-NMR, and FT-IR, which showed good agreement with the structures proposed.
2.6 Degree of branching of hyperbranched polyurethanes Equations 2.1 and 2.2 show how the degree of branching can be determined if at least one type of branching unit can be quantified. For our hyperbranched polymers 7 and 8, this is not possible by 1H-NMR, due to overlap of the signals attributed to the different branching units. Therefore, we tried to obtain this information from 13C-NMR. This analysis was only successful for hyperbranched polymer 7, since for hyperbranched polymer 8 the chemical difference between the three branching units is not sufficient to differentiate between them. In order to identify the branching units by 13C-NMR, three model compounds were synthesized that represent the three different types of units present in the final structure (figure 2.5).
O T O N H OH O O D O N H O N C6H13 H O L O N H O N C6H13 H N OH O N O O N C6H13 H N OH
Figure 2.5: Model compounds for branching units: Terminal, Dendritic, and Linear. Model compound T (terminal) was prepared by reaction of monomer 5 with di-tertbutyl dicarbonate. Reaction of T with two equivalents of hexyl isocyanate gave D (dendritic), a model compound for a dendritic subunit in which both alcohol end groups have reacted. Reaction of T with only one equivalent of hexyl isocyanate yielded a mixture of products T, D and L (linear), in a ratio of 12%:13%:75%, respectively. This composition deviates from 27
Chapter 2
the statistical distribution in favour of the formation of the linear unit. L was separated from this mixture by column chromatography. Detailed NMR analysis allowed for unambiguous assignment of the signals.
L
L D+L D+L+T
D+L
L+T
7
64 62 60 58 (ppm) 56 54 52 50
Figure 2.6: 13C-NMR spectra of the different branching units and of hyperbranched polymer 7. Figure 2.6 shows part of the 13C-NMR spectra of the three model compounds and of hyperbranched polymer 7. The NMR measurement was performed using inverse gated decoupling with a relaxation time of d1=10 s, in order to make integration as accurate as possible. Although there is overlap for most signals corresponding to the different branching units in the hyperbranched polymer, some dispersion is observed for a peak corresponding to linear units at 56.48 ppm, and a peak from terminal units at 56.38 ppm. Deconvolution of the peaks in the spectrum gives an estimation of the relative amounts of the three different branching units, namely 16%:16%:69% for terminal, dendritic, and linear branching units, respectively. The same trend is observed with the synthesis of the model compound for the linear unit L, in which a deviation from a statistical distribution favouring L was observed. Introducing these data in eq. 2.2, yields a degree of branching of DB=0.31 0.06. This is lower than the value of 0.5 that is expected based on statistics. Accordingly, with polymer 7 we have an example of a hyperbranched polymer containing predominantly linear units. Obviously, the reason for this is the lower reactivity of the second alcohol group of the monomer compared to the first one. Potentially, there are several causes for this lower reactivity. Hydrogen bonding between this alcohol group and the neighboring carbamate or 28
tertiary amine functionality lowers its nucleophilicity. Also the steric demand of the catalyst, Zr(acac)4 may be responsible for the reduced nucleophilicity. Indeed, hard metal complexes are known to be very susceptible to steric hindrance, since they act exclusively via a Lewisacid mechanism.27 Until now, no attempts have been done to increase the degree of branching of the hyperbranched polyurethanes, e.g. by using hydrogen bond competing solvents. It was impossible to determine the degree of branching of hyperbranched polymer 8, due to extensive overlap of NMR signals. Nevertheless, the degree of branching of 8 is possibly higher than that of 7, since monomer 6 is less sterically crowded.
2.7 Conclusions A facile and mild synthetic procedure for the preparation of isocyanates is described, starting from a variety of amines and di-tert-butyl tricarbonate. This procedure was used to synthesize -isocyanato--alcohols 3 from -amino--alcohols 2. By addition of a catalyst, the isocyanato alcohols were subsequently polymerized to yield the aliphatic [n]-polyurethanes. This general class of polymers was still unknown, despite its structural simplicity and strong resemblance with the [n]-nylons. The [n]-polyurethanes are highly crystalline polymers, and show a strong alternation in their melting points with respect to the number of methylene groups in the monomer unit. The polymers with an even number of methylene units in the main chain, are higher melting than the odd ones. This effect is known for many other polymers and organic compounds. Since these polymers are brittle materials, and therefore inferior to the analogous nylons, their mechanical properties were not subjected to further investigation. Using the same synthetic procedure, also aliphatic hyperbranched polyurethanes were prepared, starting from AB2 and A2B monomers. These polymers were obtained in high yields as viscous oils. A detailed analysis of hyperbranched polyurethane 7 showed that its degree of branching is quite low, DB=0.31 0.06, indicating that the hyperbranched polymer resembles linear polymers, more than dendrimers. The low degree of branching is caused by a lower reactivity of the second alcohol group of the monomer. A different catalyst, polymerization at higher temperatures, or reaction in a hydrogen bond competing solvent may lead to a higher degree of branching. It was not possible to determine the degree of branching of hyperbranched polymer 8, due to extensive overlap of NMR signals. Finally, this study shows that application of new synthetic methodologies from organic chemistry in the area of polymer synthesis is highly favorable. In the following chapters, the synthetic procedure using di-tert-butyl tricarbonate will be utilized to prepare more complex polymeric architectures, such as block copolymers. 29
Chapter 2
2.8 Experimental Section

Materials. Di-tert-butyl tricarbonate was prepared according to a literature procedure14. 2-Amino-ethanol, 3amino-propanol, 4-amino-butanol, 6-amino-hexanol, 8-amino-octanoic acid, 9-bromo-nonanol, 10-bromodecanol, 11-bromo-undecanol, laurolactam, hexylisocyanate, potassium phtalimide, and dibutyltin dilaurate were purchased from Aldrich Chemical Co, 5-amino-pentanol, lithium aluminum hydride, acrylonitrile, thionylchloride, di-tert-butyl dicarbonate, and m-cresol from Acros, oenantholactam, hydrazine hydrate, diethanolamine, and zirconium(IV) acetylacetonate, from Fluka, sodiumhydroxide and sodiumsulfate from Merck, and tetrahydrofuran, acetonitrile, methanol, chloroform, dichloromethane, and diethylether from Biosolve. Raney/cobalt was kindly provided by DSM. The amino alcohols were dried over P2O5; THF was distilled over potassium and sodium; chloroform and acetonitrile was dried over molsieves. All polymerizations were carried out under a dry argon atmosphere. Instrumentation. NMR spectra were recorded on a Varian Inova 500 MHz spectrometer, a Bruker 400 MHz spectrometer, and a Varian Gemini 300 MHz spectrometer. For the characterization of the hyperbranched polyurethanes and model compounds with 13C-NMR, the relaxation time d1=10 s, decoupler mode dm=nny, Fouries number fn=262144, number of points np=100000, and number of transients nt=8192. Infrared spectra were measured on a Perkin Elmer 1600 FT-IR. Elemental analyses were carried out using a Perkin Elmer 240. Size exclusion chromatography (SEC) was performed on a Shimadzu LC10-AT, using a Polymer Laboratories Plgel 5m Mixed-D column, a Shimadzu RID-6A detector, and N-methyl-pyrrolidone as eluent. Molecular weights were calculated relative to polystyrene standards. Differential scanning calorimetry (DSC) was performed on a Perkin DSC 7, at a heating rate of 10C min1. Thermogravimetric analysis (TGA) were performed with a Perkin-Elmer TGA 7, samples were heated from 50C to 300C at 10C min1. Solution viscosities were measured with a Schott-Gerte Ubbelohde micro-viscometer with a suspended level bulb of type 538/20: K=0.1 and 538/10: K=0.01. The sample was thermostated with a bath of type CT1450, and elution times were measured with a Schott AVS 350. m-Cresol was distilled prior to use. 7-Amino-heptanoic acid.hydrochloride Oenantholactam (4.00 g, 31.4 mmol) and concentrated hydrochloric acid (50 ml 10M in water, 0.50 mol) were heated under reflux for 3 days. The solution was allowed to cool to room temperature, active charcoal (2 g) was added and the solution was filtered over diatomaceous earth. Subsequently, the solution was evaporated to dryness under reduced pressure and the product was obtained as a slightly yellow solid (5.53 g, 97%). 1H-NMR (400 MHz, Methanol-d4): 2.92 (t, 2H, CH2NH3), 2.32 (m, 2H, CH2COOH), 1.64 (m, 4H, CH2CH2NH3 + CH2CH2COOH), 1.39 (m, 4H, CH2CH2CH2CH2CH2). Ethyl 7-amino-heptanoate.hydrochloride Thionyl chloride (4.49 g; 37.4 mmol) was added dropwise to ethanol (100 ml) cooled on an icebath. The solution was stirred for 10 min and 7-amino-heptanoic acid.hydrochloride (5.53 g, 30.4 mmol) in ethanol (40 ml) was added dropwise, stirred for another 30 min, and heated under relux for 2 h. The solution was evaporated to dryness under reduced pressure, and the product was obtained as a yellow solid (5.73 g, 90%). 1H-NMR (400 MHz, Methanol-d4): 4.10 (q, 2H, CH2O), 2.91 (t, 2H, CH2NH3), 2.33 (t, 2H, CH2COOEt), 1.64 (m, 4H, CH2CH2NH3 + CH2CH2COOEt), 1.40 (m, 4H, CH2CH2CH2CH2CH2), 1.24 (t, 3H, CH3). 7-Amino-1-heptanol, 2-(7) Ethyl 7-amino-heptanoate.hydrochloride (5.73 g; 27.3 mmol) was dissolved in dichloromethane (100 ml) and extracted with sodium hydroxide solution (60 ml 1 M). The aqueous layer was extracted two more times with dichloromethane (50 ml), and the combined organic layers were combined, dried with sodium sulfate and concentrated in vacuo. Lithium aluminumhydride (1.12 g, 29.5 mmol) was suspended in dry THF (100 ml) and cooled to 0C. To this suspension ethyl 7-amino-heptanoate in dry THF (50 ml) was added dropwise, and the solution was stirred for 30 min at 0C under an argon atmosphere, and subsequently heated under reflux for 2 h. The reaction mixture
30
was cooled to 0C, and water (2.5 ml) was added slowly. The reaction mixture was stirred overnight at 20C, filtered, dried with sodium sulfate, and concentrated under reduced pressure. The waxy solid was distilled at 110C and 0.4 mbar, yielding a white solid (0.97 g, 27%). Tm=54C (Lit: 4853C). 1H-NMR (CDCl3, 400MHz): 3.63 (t, 2H, CH2OH, J=6.6 Hz), 2.68 (t, 2H, CH2NH2, J=7.0 Hz), 1.55 (m, 2H, CH2CH2OH), 1.50 1.32 (br.m, 11H, CH2CH2NH2 + CH2CH2CH2CH2CH2 + NH2 + OH). 13C-NMR (CDCl3, 100MHz): 62.32 (CH2OH), 42.04 (CH2NH2), 33.54, 32.70, 29.20, 26.77, 25.72. Ethyl 8-amino-octanoate.hydrochloride 8-Amino-octanoic acid (2.00 g, 12.6 mmol) was dissolved in hydrochloric acid (30 ml 6 M) and the solution was evaporated to dryness under reduced pressure. The same synthetic procedure was followed as described for the synthesis of ethyl 7-amino-heptanoate.hydrochloride. The product was obtained as a yellow solid (2.68 g, 95%). 1H-NMR (400 MHz, Methanol-d4): 4.12 (q, 2H, CH2O), 2.90 (t, 2H, CH2NH3), 2.29 (t, 2H, CH2COOEt), 1.62 (m, 2H, CH2CH2NH3), 1.43 (q, 2H, CH2CH2COOEt), 1.32 (m, 6H, CH2CH2CH2CH2CH2), 1.25 (t, 3H, CH3). 8-Amino-1-octanol, 2-(8) The same synthetic procedure was followed as described for the synthesis of 7-amino-1-heptanol, starting with ethyl 8-amino-octanoate.hydrochloride (2.60 g, 11.6 mmol). The waxy solid was distilled at 100C and 0.8 mbar, yielding a white solid (0.61 g, 36%). Tm=58C (Lit: 54 59C). 1H-NMR (CDCl3, 400MHz): 3.62 (t, 2H, CH2OH, J=6.6 Hz), 2.68 (t, 2H, CH2NH2, J=7.0 Hz), 1.56 (m, 5H, CH2CH2OH + NH2 + OH), 1.44 (m, 2H, CH2CH2NH2), 1.32 (m, 8H, CH2CH2CH2CH2CH2). 13C-NMR (CDCl3, 100MHz): 62.62 (CH2OH), 42.07 (CH2NH2), 33.57, 32.73, 29.36, 29.32, 26.72, 25.66. 9-Phtalimido-1-nonanol 9-Bromo-1-nonanol (2.00 g, 8.96 mmol) and potassium phtalimide (5.11 g, 27.59 mmol) were dissolved in dry acetonitrile (70 ml) and heated under reflux for 27 h. After removal of the solvent under reduced pressure, the solid was suspended in sodium hydroxide solution (40 ml 1M) and extracted with dichloromethane (2 times 30 ml). The combined organic layers were dried with magnesium sulfate and concentrated to dryness. The product was obtained as a beige solid (2.53 g, 98%). 1H-NMR (CDCl3, 300 MHz): 7.85 (d, 2H, H3,6 arom), 7.71 (d, 2H, H4,5 arom), 3.703.63 (m, 4H, CH2N + CH2OH), 1.67 (m, 2H, CH2CH2OH), 1.58 (m, 2H, CH2CH2N), 1.30 (m, 11H, CH2CH2CH2CH2CH2 + OH). 9-Amino-1-nonanol, 2-(9) 9-Phtalimido-1-nonanol (1.30 g, 4.49 mmol) and hydrazine hydrate (0.67 g, 13.5 mmol) were dissolved in ethanol (40 ml) and stirred at 65C overnight. To the turbid reaction mixture, concentrated hydrochloric acid (1 ml) was added and it was stirred at 65C for another 2 h. After filtration, the reaction mixture was concentrated in vacuo, sodium hydroxide solution (30 ml 1M) was added, and this was extracted two times with dichloromethane (30 ml). The combined organic layers were dried with sodium sulfate, and concentrated to dryness. The solid residue was distilled at 100C and 0.5 mbar, yielding a white solid (0.44 g, 62%). Tm=69C (Lit: unknown). 1H-NMR (CDCl3, 300MHz): 3.60 (t, 2H, CH2OH, J=6.6 Hz), 2.67 (t, 2H, CH2NH2, J=7.0 Hz), 1.7 (br.s, NH2 + OH), 1.55 (m, 2H, CH2CH2OH), 1.43 (m, 2H, CH2CH2NH2), 1.30 (m, 10H, CH2CH2CH2CH2CH2). 10-Phtalimido-1-decanol The same synthetic procedure was followed as described for the synthesis of 9-phtalimido-1-nonanol, starting with 10-bromo-1-decanol (1.20 g, 5.19 mmol). The product was obtained as a white solid (1.43 g, 91%). 1HNMR (CDCl3, 400 MHz): 7.84 (d, 2H, H3,6 arom), 7.71 (d, 2H, H4,5 arom), 3.693.61 (m, 4H, CH2N + CH2OH), 1.67 (m, 2H, CH2CH2OH), 1.56 (m, 2H, CH2CH2N), 1.50 (s, 1H, OH), 1.30 (m, 12H, CH2CH2CH2CH2CH2).
31
Chapter 2
10-Amino-1-decanol, 2-(10) The same synthetic procedure was followed as described for the synthesis of 9-amino-1-nonanol, starting with 10-phtalimido-1-decanol (1.30 g, 4.28 mmol). The solid residue was distilled at 125C and 0.6 mbar, yielding a white solid (0.47 g, 63%). Tm=72C (Lit: 72C). 1H-NMR (CDCl3, 300MHz): 3.63 (t, 2H, CH2OH, J=6.6 Hz), 2.67 (t, 2H, CH2NH2, J=6.9 Hz), 1.56 (m, 2H, CH2CH2OH), 1.43 (m, 2H, CH2CH2NH2), 1.29 (m, 15H, CH2CH2CH2CH2CH2 + NH2 + OH). 11-Phtalimido-1-undecanol The same synthetic procedure was followed as described for the synthesis of 9-phtalimido-1-nonanol, starting with 11-bromo-1-undecanol (1.00 g, 4.25 mmol). The product was obtained as a white solid (1.24 g, 92%). 1HNMR (CDCl3, 400 MHz): 7.84 (d, 2H, H3,6 arom), 7.71 (d, 2H, H4,5 arom), 3.693.62 (m, 4H, CH2N + CH2OH), 1.67 (m, 2H, CH2CH2OH), 1.54 (m, 2H, CH2CH2N), 1.321.37 (m, 15H, CH2CH2CH2CH2CH2+ OH). 11-Amino-1-undecanol, 2-(11) The same synthetic procedure was followed as described for the synthesis of 9-amino-1-nonanol, starting with 11-phtalimido-1-undecanol (0.58 g, 1.83 mmol). The white solid product was not distilled (0.27 g, 79%). Tm=85C (Lit: 73C). 1H-NMR (CDCl3, 400MHz): 3.64 (t, 2H, CH2OH, J=6.6 Hz), 2.68 (t, 2H, CH2NH2, J=7.0 Hz), 1.56 (m, 2H, CH2CH2OH), 1.43 (m, 2H, CH2CH2NH2), 1.28 (m, 17H, CH2CH2CH2CH2CH2 + NH2 + OH). 13C-NMR (CDCl3, 100MHz): 62.98 (CH2OH), 42.27 (CH2NH2), 33.86, 32.80, 29.56, 29.48 (m), 29.39, 26.86, 25.72. 12-Amino-dodecanoic acid.hydrochloride Laurolactam (10.00 g, 50.7 mmol) and concentrated hydrochloric acid (200 ml 10M in water, 2.00 mol) were heated under reflux for 4 days. The product crystallized by cooling the solution to 0C, was filtered off, washed with water, and dried in vacuo. The product was obtained as colorless flakes (11.50 g, 90%). 1H-NMR (400 MHz, Methanol-d4): 2.91 (t, 2H, CH2NH3), 2.27 (m, 2H, CH2COOH), 1.66 (m, 2H, CH2CH2NH3), 1.59 (m, 2H, CH2CH2COOH), 1.32 (br. m, 14H, CH2CH2CH2CH2CH2). Ethyl 12-amino-dodecanoate.hydrochloride The same synthetic procedure was followed as described for the synthesis of ethyl 7-aminoheptanoate.hydrochloride, starting with 12-amino-dodecanoic acid.hydrochloride (5.00 g, 19.9 mmol). The product was obtained as a white solid (5.57 g, 100%). 1H-NMR (400 MHz, Methanol-d4): 4.11 (q, 2H, CH2O), 2.90 (t, 2H, CH2NH3), 2.29 (t, 2H, CH2COOEt), 1.661.57 (m, 4H, CH2CH2NH3 + CH2CH2COOEt), 1.31 (m, 14H, CH2CH2CH2CH2CH2 + NH3), 1.24 (t, 3H, CH3). 12-Amino-1-dodecanol, 2-(12) The same synthetic procedure was followed as described for the synthesis of 7-amino-1-heptanol, starting with ethyl 12-amino-dodecanoate.hydrochloride (5.57 g, 19.9 mmol). The waxy solid was distilled at 140C and 0.2 mbar, yielding a white solid (1.52 g, 38%). Tm=77C (Lit: 79 80C). 1H-NMR (CDCl3, 400MHz): 3.64(t, 2H, CH2OH, J=6.8 Hz), 2.68 (t, 2H, CH2NH2, J=7.2 Hz), 1.56 (m, 2H, CH2CH2OH), 1.44 (m, 2H, CH2CH2NH2), 1.27 (m, 19H, CH2CH2CH2CH2CH2 + NH2 + OH). 13C-NMR (CDCl3, 100MHz): 63.00 (CH2OH), 42.27 (CH2NH2), 33.85, 32.79, 29.54, 29.45 (m), 29.38, 26.86, 25.71. [n]-Polyurethane, 4-(x) The synthesis of [6]-polyurethane, 4-(5) is given as a typical example. A solution of 5-amino-1-pentanol (9.69 mmol, 1.00 g) in chloroform (2 ml) was injected by a syringe under the surface of a stirred solution of di-tert-butyl tricarbonate (10.66 mmol, 2.80 g) in chloroform (30 ml). The solution was stirred for 10 min at room temperature under an argon atmosphere. 1H-NMR (400 MHz, CDCl3, 20C, TMS): 3.67 (q, 2H, CH2OH, J=6.0 Hz), 3.32 (t, 2H, CH2NCO, J=6.6 Hz), 1.55 (br.m, 4H, CH2CH2OH + CH2CH2NCO), 1.40 (m, 2H, CH2 CH2CH2CH2 CH2, J=6.7 Hz). IR (CHCl3): 3396 (br.s, OH), 2971 (m, C H), 2274 (s, N=C=O) cm1.
32
Zirconium(IV) acetylacetonate (0.1 mol%) was added, and the polymerization was carried out for 20 h with continuous stirring under argon at room temperature. The turbid reaction mixture was precipitated in diethyl ether (150 ml) and the polymeric product was collected by suction filtration in a yield of 0.81 g (63%). Tm=127C; Tdecomp=200C. 1H-NMR (400 MHz, DMSO-d6, 20C, TMS): 7.04 (br.t, 0.9H, NH anti conformer), 6.72 (br.m, 0.1H, NH syn conformer), 3.89 (t, 2H, CH2O, J=6.2 Hz), 2.95 (q, 2H, CH2N, J=6.0 Hz), 1.50 (m, 2H, CH2CH2O, J=7.2 Hz), 1.40 (m, 2H, CH2CH2N, J=7.7 Hz), 1.28 (m, 2H, CH2 CH2CH2CH2 CH2, J=6.6 Hz), 13C-NMR (DMSO-d6, 100 MHz, 100C): 156.4 (C=O), 63.5 (CH2O), 40.1 (CH2N), 29.4 (CH2CH2O), 28.7 (CH2CH2N), 26.0 (CH2 CH2CH2CH2 CH2). IR (KBr): 3318 (br.s, NH), 2944 (m, CH), 2870 (w), 1684 (s, C=O), 1535 (s, amide II), 1263 (s) cm1. Anal. Calcd. (%) for (C6H11NO2)n: C 55.80, H 8.58, N 10.85. Found (%): C 55.46, H 8.67, N 10.55. [5]-Polyurethane, (C4H8NHCOO)n, 4-(4) H-NMR (400 MHz, DMSO-d6, 20C, TMS): 7.10 (br.t, 0.9H; NH anti conformer), 6.70 (br.m, 0.1H; NH syn conformer), 3.90 (t, 2H; CH2O), 2.96 (q, 2H; CH2N), 1.48 (m, 2H; CH2CH2O), 1.40 (m, 2H; CH2CH2N). IR (KBr): 3317 (br.s), 2954 (m), 1689 (s), 1541 (s), 1271 (s) cm1. Anal. Calcd. (%) for (C5H9NO2)n: C 52.16, H 7.88, N 12.17. Found(%): C 51.79, H 8.03, N 11.92. Tm=194C; Tdecomp=180C.
1
[7]-Polyurethane (C6H12NHCOO)n, 4-(6) 1 H-NMR (400 MHz, DMSO-d6, 20C, TMS): 7.03 (br.t, 0.9H; NH anti conformer), 6.75 (br.m, 0.1H; NH syn conformer), 3.90 (t, 2H; CH2O), 2.94 (q, 2H; CH2N), 1.50 (m, 2H; CH2CH2O), 1.37 (m, 2H; CH2CH2N), 1.17 (m, 4H; CH2 CH2CH2CH2 CH2); IR (KBr): 3319 (br.s), 2938 (m), 2860 (w), 1687 (s), 1542 (s), 1257 (s) cm1; Anal. Calcd. (%) for (C7H13NO2)n: C 58.72, H 9.15, N 9.78. Found (%): C 58.67, H 9.21, N 9.57. Tm=157C; Tdecomp=210C. [8]-Polyurethane, (C7H14NHCOO)n, 4-(7) 1 H-NMR (400 MHz, DMSO-d6, 20C, TMS): 7.03 (br.t, 0.9H; NH anti conformer), 6.73 (br.m, 0.1H; NH syn conformer), 3.89 (t, 2H; CH2O), 2.93 (q, 2H; CH2N), 1.48 (m, 2H; CH2CH2O), 1.38 (m, 2H; CH2CH2N), 1.24 (m, 6H; CH2 CH2CH2CH2 CH2); IR (KBr): 3327 (br.s), 2934 (m), 2856 (w), 1687 (s), 1534 (s), 1252 (s) cm1; Anal. Calcd. (%) for (C8H15NO2)n: C 61.12, H 9.62, N 8.91. Found (%): C 60.66, H 10.09, N 8.65. Tm=114C. [9]-Polyurethane, (C8H16NHCOO)n, 4-(8) 1 H-NMR (400 MHz, DMSO-d6, 100C, TMS): 6.47 (br.t, 1H; NH), 3.93 (t, 2H; CH2O), 2.97 (q, 2H; CH2N), 1.51 (m, 2H; CH2CH2O), 1.40 (m, 2H; CH2CH2N), 1.27 (m, 8H; CH2 CH2CH2CH2 CH2); IR (KBr): 3321 (br.s), 2928 (m), 2852 (w), 1686 (s), 1545 (s), 1261 (s) cm1; Anal. Calcd. (%) for (C9H17NO2)n: C 63.13, H 10.01, N 8.18. Found (%): C 62.84, H 10.64, N 7.88. Tm=146C. [10]-Polyurethane, (C9H18NHCOO)n, 4-(9) 1 H-NMR (400 MHz, DMSO-d6, 120C, TMS): 6.43 (br.t, 1H; NH), 3.95 (t, 2H; CH2O), 3.00 (q, 2H; CH2N), 1.56 (m, 2H; CH2CH2O), 1.43 (m, 2H; CH2CH2N), 1.28 (m, 10H; CH2 CH2CH2CH2 CH2); IR (KBr): 3325 (br.s), 2925 (m), 2852 (w), 1686 (s), 1535 (s), 1254 (s) cm1; Anal. Calcd. (%) for (C10H19NO2)n: C 64.83, H 10.34, N 7.56. Found (%): C 64.74, H 10.79, N 7.33. Tm=116C. [11]-Polyurethane, (C10H20NHCOO)n, 4-(10) 1 H-NMR (400 MHz, DMSO-d6, 120C, TMS): (br.t, 1H; NH), 3.95 (t, 2H; CH2O), 3.00 (q, 2H; CH2N), 1.55 (m, 2H; CH2CH2O), 1.43 (m, 2H; CH2CH2N), 1.28 (m, 12H; CH2 CH2CH2CH2 CH2); IR (KBr): 3324 (br.s), 2922 (m), 2851 (w), 1687 (s), 1543 (s), 1250 (s) cm1; Anal. Calcd. (%) for (C11H21NO2)n: C 66.30, H 10.62, N 7.03. Found (%): C 65.86, H 11.01, N 6.82. Tm=148C. [12]-Polyurethane, (C11H22NHCOO)n, 4-(11) 1 H-NMR (400 MHz, DMSO-d6, 100C, TMS): 6.40 (br.t, 1H; NH), 3.92 (t, 2H; CH2O), 2.97 (q, 2H; CH2N), 1.51 (m, 2H; CH2CH2O), 1.40 (m, 2H; CH2CH2N), 1.28 (m, 14H; CH2 CH2CH2CH2 CH2); IR (KBr): 3327
33
Chapter 2
(br.s), 2922 (m), 2851 (w), 1684 (s), 1534 (s), 1246 (s) cm1; Anal. Calcd. (%) for (C12H23NO2)n: C 67.57, H 10.87, N 6.57. Found (%): C 66.98, H 11.16, N 6.22. Tm=119C. [13]-Polyurethane, (C12H24NHCOO)n, 4-(12) 1 H-NMR (400 MHz, DMSO-d6, 100C, TMS): 6.38 (br.t, 1H; NH), 3.92 (t, 2H; CH2O), 2.95 (q, 2H; CH2N), 1.51 (m, 2H; CH2CH2O), 1.40 (m, 2H; CH2CH2N), 1.25 (m, 16H; CH2 CH2CH2CH2 CH2); IR (KBr): 3326 (br.s), 2921 (m), 2850 (w), 1686 (s), 1542 (s), 1242 (s) cm1; Anal. Calcd. (%) for (C13H25NO2)n: C 68.68, H 11.08, N 6.16. Found (%): C 68.25, H 11.40, N 5.93. Tm=141C. [6-co-7]-Polyurethane, (C5H10NHCOO-co-C6H12NHCOO)n 1 H-NMR (400 MHz, DMSO-d6, 20C, TMS): 7.05 (br.t, 0.9H; NH anti conformer), 6.73 (br.m, 0.1H; NH syn conformer), 3.90 (t, 2H; CH2O), 2.94 (q, 2H; CH2N), 1.50 (m, 2H; CH2CH2O), 1.38 (m, 2H; CH2CH2N), 1.27 (m, 3H; CH2 CH2CH2CH2 CH2); IR (KBr): 3324 (br.s), 2941 (m), 2850 (w), 1687 (s), 1540 (s), 1259 (s) cm1; Anal. Calcd. (%) for (C5.5H11NO2)n: C 57.26, H 8.87, N 10.32. Found (%): C 56.84, H 8.94, N 9.98. Tg=8C. N,N-Bis-(2-hydroxyethyl)-amino-propionitrile Diethanolamine (20 g, 0.19 mol) was dissolved in water (100 ml), and cooled on an icebath. Acrylonitrile (53 g, 1.00 mol) was added dropwise, and the reaction mixture was stirred for 1h at 0C, and subsequently heated to 70C for 3 h. After evaporation of water and excess arylonitrile under reduced pressure, the product was obtained as colorless liquid (29.16 g, 97%). 1H-NMR (CDCl3, 400 MHz): 3.85 (br.s, 2H; OH), 3.61 (t, 4H, CH2OH, J=5.2 Hz), 2.91 (t, 2H, NCH2CH2CN, J=6.8 Hz), 2.69 (t, 4H, NCH2CH2OH, J=5.2 Hz), 2.55 (t, 2H, CH2CN, J=6.8 Hz). 13C-NMR (CDCl3,100 MHz): 119.22 (CN), 60.71 (CH2OH), 55.75 (CH2CH2OH), 50.07 (NCH2CH2CN), 16.25 (CH2CN). FT-IR (neat): 3380 (br.s, OH), 2950 (m, CH), 2248 (m, CN), 1642 (m), 1027 (s, CO) cm1. N,N-Bis-(2-hydroxyethyl)-propylenediamine, 5 Raney/cobalt catalyst suspension in water (3 g) was decanted, rinsed two times with methanol and added to a methanolic ammonia solution (100 ml 7 N). N,N-bis-(2-hydroxyethyl)-amino-propionitrile (26 g, 0.16 mol) was added to this suspension, and the reaction mixture was transferred into the Parr-reactor vessel. After closing the reactor, the solution was purged three times with nitrogen, and two times with hydrogen gas. The reaction mixture was mechanically stirred during 2 h at 50C and 85 bar hydrogen pressure. After cooling and releasing pressure, the catalyst was filtered off on a glass filter over a layer of diatomaceous earth, and the reaction mixture was evaporated under reduced pressure until dryness. The brown product was distilled at 180C and 0.05 mbar using a Kugelrohr apparatus. The product was obtained as a colorless liquid (22.11 g, 85%). 1H-NMR (CDCl3, 400 MHz): 3.62 (t, 4H, CH2OH, J=5.4 Hz), 2.84 (t, 2H, CH2NH2, J=6.4 Hz), 2.8 (br. s, 4H, OH + NH2), 2.65 (t, 2H, CH2CH2CH2NH2, J=6.4 Hz), 2.61 (t, 4H, CH2CH2OH, J=5.2 Hz), 1.62 (qui, 2H, CH2CH2CH2, J=6.4 Hz). 13C-NMR (CDCl3, 100 MHz): 59.66 (CH2OH), 56.09 (CH2CH2OH), 52.20 (CH2CH2CH2NH2), 40.13 (CH2NH2), 29.51 (CH2CH2CH2). FT-IR (neat): 3345 and 3289 (br.s, OH + NH2), 2939 (s, CH), 1600 (m), 1036 (s, CO) cm1. N,N-Bis-(2-cyanoethyl)-3-amino-1-propanol The same synthetic procedure was followed as described for the synthesis of N,N-bis-(2-hydroxyethyl)-aminopropionitrile, starting with 3-amino-1-propanol (10.00 g, 0.133 mol). The product was obtained as a colorless oil (24.09 g, 100%). 1H-NMR (CDCl3, 400 MHz): 3.75 (t, 2H, CH2OH), 2.89 (t, 4H, NCH2CH2CN), 2.70 (t, 2H, NCH2CH2CH2OH), 2.60 (br.s, 1H, OH), 2.51 (t, 4H, CH2CN), 1.72 (qui, 2H, CH2CH2CH2). 13C-NMR (CDCl3, 100 MHz): 118.64 (CN), 60.71 (CH2OH), 50.65 (NCH2CH2CH2OH), 49.48 (NCH2CH2CN), 29.45 (CH2CH2CH2) 16.70 (CH2CN). FT-IR (neat): 3424 (br.s, OH), 2950 (m, CH), 2248 (s, CN), 1466 (m), 1421 (m), 1367 (m), 1033 (s, CO) cm1.
34
N,N-Bis-(3-aminopropyl)-3-amino-1-propanol, 6 The same synthetic procedure was followed as described for the synthesis of N,N-bis-(2-hydroxyethyl)propylenediamine, starting with N,N-bis-(2-cyanoethyl)-3-amino-1-propanol (23.33 g, 0.129 mol). The product was obtained as a colorless liquid (20.10 g, 80%). 1H-NMR (CDCl3, 400 MHz): 3.70 (t, 2H, CH2OH, J=5.6 Hz), 2.72 (t, 4H, CH2NH2, J=6.8 Hz), 2.60 (t, 2H, CH2CH2CH2OH, J=6.2 Hz), 2.48 (t, 4H, CH2CH2CH2NH2, J=7.4 Hz), 2.24 (br. s, 5H, OH + NH2), 1.69 (qui, 2H, OCH2CH2CH2N, J=5.8 Hz). 1.62 (m, 4H, NCH2CH2CH2N, J=7.2 Hz). 13C-NMR (CDCl3, 100 MHz): 62.25 (CH2OH), 53.02 (CH2CH2CH2OH), 51.28 (CH2CH2CH2NH2), 39.98 (CH2NH2), 30.17 (NCH2CH2CH2N), 28.14 (OCH2CH2CH2N). FT-IR (neat): 3340 and 3284 (br.s, OH + NH2), 2942 (s, CH), 1599 (m), 1039 (s, CO) cm1. Hyperbranched polymer, 7 In a solution of di-tert-butyl tricarbonate (0.93 g, 3.56 mmol) in dry chloroform (10 ml) was quickly injected a solution of N,N-bis-(2-hydroxyethyl)-propylenediamine (0.55 g, 3.39 mmol) in dry chloroform (2 ml). The solution was stirred for 15 min, zirconium(IV) acetylacetonate (0.1 mol%) was added, and the polymerization was carried out for 20 h with continuous stirring under argon at room temperature. The reaction mixture was evaporated to dryness under reduced pressure. The product was obtained as a yellow viscous oil (0.60 g, 94%). 1 H-NMR (DMSO-d6, 500 MHz): 7.02 (1H, NH), 4.34 (1H, OH), 3.96 (2H, CH2OC=O) 3.43 (2H, CH2OH), 2.99 (2H, CH2NH,), 2.64 (2H, CH2CH2CH2NH), 2.50 (4H, CH2CH2O), 1.51 (2H, CH2CH2CH2). 13C-NMR (DMSO-d6, 125 MHz, d1=10): 156.20 (CH2OC=O), 61.86 (CH2OC=O), 59.26 (CH2OH), 56.48 (NCH2CH2OH), 53.06 (NCH2CH2OC=O), 52.11 (NHCH2CH2CH2N), 38.45 (NHCH2CH2CH2N), 27,13 (CH2CH2CH2). FT-IR (ATR): 3312 (br. s, NH + OH), 2946 (m, CH), 1694 (br. s, C=O), 1543 (s, amide II), 1250 (s), 1038 (s) cm-1. SEC (NMP, rel. to PS): 6.4*103 g/mol. Hyperbranched polymer, 8 In a solution of di-tert-butyl tricarbonate (1.40 g, 5.34 mmol) in dry chloroform (28 ml) was quickly injected a solution of N,N-bis-(3-aminopropyl)-3-amino-1-propanol (0.50 g, 2.64 mmol) in dry chloroform (2 ml). The solution was stirred for 15 min, and subsequently the solvent and tert-butanol were evaporated under reduced pressure. The residue was redissolved in dry chloroform (30 ml) and zirconium(IV) acetylacetonate (0.1 mol%) was added, and the polymerization was carried out for 20 h with continuous stirring under argon at room temperature. Next, methanol (20 ml) was added, and the reaction mixture was stirred for another 8 h, after which it was partly concentrated. Precipitation in diethyl ether resulted in a turbid emulsion, which was centrifugated and decanted, and the viscous residue was dried in vacuo. The product was obtained as a yellow viscous oil (0.37 g, 60%). 1H-NMR (DMSO-d6, 400 MHz): 7.06 (2H, NH), 3.92 (2H, CH2O), 3.50 (3H, CH3O), 2.97 (4H, CH2NH), 2.33 (6H, N(CH2)3), 1.61 (2H, OCH2CH2CH2), 1.48 (4H, NCH2CH2CH2N). 13C-NMR (DMSO-d6, 100 MHz): 157.22 (CH2OC=O), 156.97 (CH3OC=O), 62.75 (CH2O), 51.89 (N(CH2)3), 50.15 (CH3O), 40.25 (CH2NH), 26.91 (NCH2CH2CH2N), 26.49 (OCH2CH2CH2N). FT-IR (ATR): 3333 (br. s, NH), 2971 (m, C H), 1699 (br. s, C=O), 1530 (s, amide II) cm-1. SEC (NMP, rel. to PS): 3.8*103 g/mol. N-[3-(t-butoxycarbonylamino)propyl]-diethanolamine, model compound terminal unit, T Di-tert-butyl dicarbonate (4.04 g, 18.5 mmol) was dissolved in chloroform (20 ml), and N,N-bis-(2hydroxyethyl)-propylenediamine (3.00 g, 18.5 mmol) in chloroform (5 ml) was added. The reaction mixture was stirred for 3 h at room temperature, and was then evaporated to dryness under reduced pressure. The product was obtained as a colorless liquid (4.85 g, 100%). 1H-NMR (DMSO-d6, 400 MHz): 6.74 (t, 1H, NH), 4.29 (s, 2H, OH), 3.41 (t, 4H, CH2OH, J=6.4 Hz), 2.92 (q, 2H, NHCH2, J=6.4 Hz), 2.48 (t, 4H, NCH2CH2OH, J=6.4 Hz), 2.43 (t, 2H, NHCH2CH2CH2N, J=7.2 Hz), 1.49 (qui, 2H, CH2CH2CH2, J=6.8 Hz), 1.37 (s, 9H, CH3). 13C-NMR (DMSO-d6, 125 MHz, d1=10): 155.61 (1C, C=O), 77.29 (1C, C(CH3)), 59.05 (2C, CH2OH), 56.38 (2C, NCH2CH2OH), 52.09 (1C, NHCH2CH2CH2N), 37.99 (1C, NHCH2), 28.08 (3C, CH3), 26.99 (1C, CH2CH2CH2). FT-IR (ATR): 3336 (br. s, NH + OH), 2936 (m, CH), 1686 (s, C=O), 1525 (s, amide II), 1166 (s), 1039 (s) cm-1.
35
Chapter 2
model N-[3-(t-butoxycarbonylamino)propyl]-N-[2-(hexylaminocarbonyloxyethyl)]-ethanolamine, compound dendritic unit, D N,N-Bis-(2-hydroxyethyl)-propylenediamine (1.00 g, 3.82 mmol) and hexyl isocyanate (0.97 g, 7.62 mmol) were dissolved in chloroform (10 ml), and zirconium(IV) acetylacetonate (0.1 mol%) was added. The reaction mixture was stirred for 2 days at 20C under an argon atmosphere. and was then evaporated to dryness under reduced pressure. The product was obtained as a colorless liquid (1.95 g, 99%). 1H-NMR (DMSO-d6, 400 MHz): 7.02 (t, 2H, CH2OC=ONH), 6.74 (t, 1H, (CH3)3COC=ONH), 3.94 (t, 4H, CH2OC=O, J=5.8 Hz), 2.94 (m, 6H, CH2NH, J=6.6 Hz), 2.62 (t, 4H, NCH2CH2OC=O), 2.46 (t, 2H, NHCH2CH2CH2N), 1.47 (qui, 2H, NHCH2CH2CH2N), 1.37 (m, 13H, C(CH3)3 + NHCH2CH2C4H9), 1.24 (m, 12H, (CH2)3CH3), 0.85 (t, 6H, CH2CH3, J=6.8 Hz). 13C-NMR (DMSO-d6, 125 MHz, d1=10): 156.16 (2C, CH2OC=O), 155.61 (1C, (CH3)3COC=ONH), 77.31 (1C, C(CH3)), 61.82 (2C, CH2OC=O), 52.85 (2C, NCH2CH2OC=O), 51.91 (1C, NHCH2CH2CH2N), 40.23 (2C, NHCH2C5H11), 38.11 (1C, NHCH2CH2CH2N), 31.05 (2C, CH2), 29.44 (2C, CH2), 28.17 (3C, C(CH3)3), 27.19 (1C, NHCH2CH2CH2N), 25.98 (2C, CH2), 22.12 (2C, CH2), 13.86 (1C, CH3). FT-IR (ATR): 3325 (br. s, NH), 2932 (m, CH), 1708 (s, C=O), 1518 (s, amide II), 1216 (s) cm-1. N-[3-(t-butoxycarbonylamino)propyl]-N,N-bis-[2-(hexylaminocarbonyloxyethyl)]amine, model compound linear unit, L N,N-Bis-(2-hydroxyethyl)-propylenediamine (1.00 g, 3.82 mmol) and hexyl isocyanate (0.48 g, 3.82 mmol) were dissolved in chloroform (10 ml), and zirconium(IV) acetylacetonate (0.1 mol%) was added. The reaction mixture was stirred for 1 day at 20C under an argon atmosphere. and was then evaporated to dryness under reduced pressure. The product was purified by column chromatography using 4% methanol in chloroform (Rf=0.3). The product was obtained as a colorless liquid (1.04 g, 70%). 1H-NMR (DMSO-d6, 400 MHz): 6.98 (t, 1H, CH2OC=ONH), 6.70 (t, 1H, (CH3)3COC=ONH), 4.22 (s, 1H, OH), 3.93 (t, 2H, CH2OC=O, J=6 Hz), 3.39 (t, 2H, CH2OH), 2.92 (m, 4H, CH2NH, J=6.8 Hz), 2.58 (t, 2H, NCH2CH2OC=O, J=6 Hz), 2.48 (t, 2H, NCH2CH2OH, J=6.2 Hz), 2.42 (t, 2H, NHCH2CH2CH2N, J=6.6 Hz), 1.45 (qui, 2H, NHCH2CH2CH2N, J=6.6 Hz), 1.35 (m, 11H, C(CH3)3 + NHCH2CH2C4H9), 1.21 (m, 6H, (CH2)3CH3), 0.83 (t, 3H, CH2CH3, J=6.6 Hz). 13C-NMR (DMSO-d6, 125 MHz, d1=10): 156.21 (1C, CH2OC=O), 155.58 (1C, (CH3)3COC=ONH), 77.27 (1C, C(CH3)), 61.79 (1C, CH2OC=O), 59.26 (1C, CH2OH), 56.49 (1C, NCH2CH2OH), 53.03 (1C, NCH2CH2OC=O), 52.09 (1C, NHCH2CH2CH2N), 40.19 (1C, NHCH2C5H11), 38.12 (1C, NHCH2CH2CH2N), 30.99 (1C, CH2), 29.39 (1C, CH2), 28.12 (3C, C(CH3)3), 27.07 (1C, NHCH2CH2CH2N), 25.91 (1C, CH2), 22.05 (1C, CH2), 13.86 (1C, CH3). FT-IR (ATR): 3331 (br. s, NH + OH), 2931 (m, CH), 1689 (s, C=O), 1525 (s, amide II), 1249 (s), 1168 (s), 1043 (m) cm-1.

1. a) Wirpsza, Z. Polyurethanes: Chemistry, Technology and Applications; Ellis Horwood: London, 1993; b) Mark, H. F. Encyclopedia of polymer science and technology; 3 ed.; Wiley-Interscience: New York, 2001; c) Saunders, J. H. and Frisch, K. C. Polyurethanes : chemistry and technology, part 1 Chemistry; Interscience: New York, 1962; Vol. 16; d) Saunders, J. H. and Frisch, K. C. Polyurethanes : chemistry and technology, part 2 Technology; Interscience: New York, 1964; Vol. 16; e) David, D. J. and Stanley, H. B. Analytical chemistry of polyurethanes, part 3; Wiley-Interscience: New York, 1969; Vol. 16. 2. a) Spaans, C. J. Ph.D. Thesis, University of Groningen (Groningen), 2000; b) Runt, J., Xu, R., Manias, E. and Snyder, A. J. Macromolecules 2001, 34, 337. 3. Bayer, O. Angew. Chem. 1947, 59, 257. 4. Bachmann, F., Reimer, J., Ruppenstein, M. and Thiem, J. Macromol. Rapid Commun. 1998, 19, 21. 5. a) Carothers, W. H. (DuPont) Patent US2130948, 1938; b) Carothers, W. H. (DuPont) Patent US2071253, 1937. 6. a) Schaefgen, J. R., Koontz, F. H. and Tietz, R. F. J. Polym. Sci. 1959, 40, 377; b) Iwakura, Y. Kobunshi Kagaku 1945, 2, 305. 7. Neffgen, S., Kusan, J., Fey, T., Keul, H. and Hocker, H. Macromol. Chem. Phys. 2000, 201, 2108.
36
8. a) Neffgen, S., Keul, H. and Hcker, H. Macromol. Rapid Commun. 1996, 17, 373; c) Neffgen, S., Keul, H. and Hcker, H. Macromolecules 1997, 30, 1289. 9. a) Iwakura, Y., Hayashi, K. and Iwata, K. Macromolecular Chemistry 1965, 89, 214; b) Schultz, P. G., Cho, C. Y., Moran, E. J., Cherry, S. R., Stephans, J. C., Fodor, S. P. A., Adams, C. L., Sundaram, A. and Jacobs, J. W. Science 1993, 261, 1303; c) Masuyama, A., Tsuchiya, K. and Okahara, M. Bull. Chem. Soc. Japan 1985, 58, 2855. 10. a) Kinstle, J. F. and Sepulveda, L. E. J. Polym. Sci. Polym. Lett. Ed. 1977, 15, 467; b) Kurita, K., Matsuda, S. and Iwakura, Y. Macromol. Rapid Commun. 1980, 1, 211; c) Ghadge, N. D. and Jadhav, J. Y. J. Polym. Sci. 1983, 21, 1941. 11. Peerlings, H. W. I. and Meijer, E. W. Tetrahedron Lett. 1999, 40, 1021. 12. Pope, B. M., Yamamoto, Y. and Tarbell, D. S. Org. Synth. 1978, 57, 45. 13. March, J. Advanced organic chemistry; 4 ed.; J. Wiley & Sons: New York, 1992. 14. Deetz, M. J., Forbes, C. C., Jonas, M., Malerich, J. P., Smith, B. D. and Wiest, O. J. Org. Chem. 2002, ASAP. 15. Aharoni, S. M. n-Nylons: Their synthesis, structure, and properties; J. Wiley & Sons: Chichester, 1997, pp 60. 16. a) Baeyer, A. Ber. Chem. Ges. 1877, 10, 1286; b) Boese, R., Weiss, H.-C. and Blaser, D. Angew. Chem. 1999, 38, 988; c) Thalladi, V. R., Boese, R. and Weiss, H.-C. Angew. Chem. 2000, 39, 918; d) Thalladi, V. R., Boese, R. and Weiss, H.-C. J. Am. Chem. Soc. 2000, 122, 1186; e) Thalladi, V. R., Nsse, M. and Boese, R. J. Am. Chem. Soc. 2000, 122, 9227. 17. Krevelen, D. W. v. Properties of Polymers; Elsevier: Amsterdam, 1997, pp 151. 18. Frchet, J. M. J., Hawker, C. J., Gitsov, I. and Leon, J. W. J. Macromol. Sci. 1996, 33, 1399. 19. Bosman, A. W., Janssen, H. M. and Meijer, E. W. Chem. Rev. 1999, 99, 1665. 20. a) Kim, Y. H. J. Polym. Sci. 1998, 36, 1685; b) Hult, A., Johansson, M. and Malmstrm, E. Adv. Polym. Sci. 1998, 143, 1; c) Sunder, A., Heinemann, J. and Frey, H. Chem. Eur. J. 2000, 6, 2499. 21. Flory, P. J. Principles of Polymer Chemistry; Cornell University Press: Ithaca: New York, 1953. 22. Hawker, C. J., Lee, R. and Frechet, J. M. J. J. Am. Chem. Soc. 1991, 113, 4583. 23. Frey, H. and Holter, D. Polym. Mater. Sci. Eng. 1997, 77, 226. 24. Markoski, L. J., Thompson, J. L. and Moore, J. S. Macromolecules 2002, 35, 1599. 25. a) Jayakannan, M. and Ramakrishnan, S. Chem. Commun. 2000; b) Kumar, A. and Meijer, E. W. Chem. Commun. 1998; c) van Benthem, R. A. T. M., Meijerink, N., Gelade, E., de Koster, C. G., Muscat, D., Froehling, P. E., Hendriks, P. H. M., Vermeulen, C. J. A. A. and Zwartkruis, T. J. G. Macromolecules 2001, 34, 3559. 26. a) Spindler, R. and Frchet, J. M. J. Macromolecules 1993, 26, 4809; b) Kumar, A. and Ramakrishnan, S. J. Chem. Soc. 1993; c) Kumar, A. and Ramakrishnan, S. J. Polym. Sci. A 1996, 34, 839; d) Clark, A. J., Echenique, J., Haddleton, D. M., Straw, T. A. and Taylor, P. C. J. Org. Chem. 2001, 66, 8687; e) Hong, L., Cui, Y., Wang, X. and Tang, X. J. Polym. Sci. A 2002, 40, 344; f) Taylor, R. T. and Puapaiboon, U. Tetrahedron Lett. 1998, 39, 8005. 27. Thiele, L. and Becker, R. Adv. Ureth. Sci. 1993, 12, 59.
37
3
Synthesis and characterization of block copoly(ether urea)s
Abstract Block copoly(ether urea)s were prepared from amine-terminated prepolymers, employing the isocyanate chemistry that was described in the previous chapter. The amine-terminated prepolymers were synthesized from commercially available hydroxy-terminated prepolymers via a two-step synthesis; a cyanoethylation (Michael addition of acrylonitrile) and subsequently a hydrogenation of the nitrile end groups. The ,-diaminopoly(tetrahydrofuran)s were obtained in excellent yields on a multigram scale. A heterofunctional prepolymer with two different end groups, amino--hydroxy-poly(tetrahydrofuran), was prepared by cationic ring opening polymerization of tetrahydrofuran, employing a strategy of functional initiation and functional termination. Due to the relatively slow initiation, the molecular weight was difficult to control, while low yields were obtained. The amine-terminated prepolymers were used as soft block in the block copoly(ether urea)s. Via protective group strategy and novel isocyanate chemistry employing di-tert-butyl tricarbonate, segmented copolymers with uniform hard blocks consisting of exactly 1 to 4 urea groups were prepared. The strength of hydrogen bonding between the urea groups in the hard block was studied by infrared spectroscopy. The polymers possessing two urea groups in the hard block, show the optimal balance between the strength of hydrogen bonding, and the processibility/solubility. They possess intriguing elastomeric properties. The nature of both the hard and soft blocks were varied, and their influence on the properties are discussed.
39
Chapter 3
3.1 Introduction 3.1.1 Thermoplastic elastomers Thermoplastic elastomers (TPEs) are polymers that combine advantages of both thermoplastics and elastomers (see chapter 1).1 In general, their specific properties are a result of their morphology. At ambient temperature, physical crosslinks in the amorphous matrix give the material its elastomeric, rubber-like properties. At higher temperatures, these physical crosslinks are broken due to their reversibility, and the material can be processed easily, characteristic for thermoplastics. In many thermoplastic elastomers, the reversible crosslinks originate from crystallization of one of the blocks of the segmented copolymer. These hard blocks are generally polyester1,2, polyamide3, or polyurethane4 segments. At low temperatures, crystalline domains account for the mechanical stability of the material; however, above the melting point of the hard block, a polymer melt is obtained.
residue of long chain diol residue of chain extender residue of diisocyanate urethane group
Figure 3.1: Schematic representation of a TPU and its building blocks.1 From the segmented copolymers, the thermoplastic elastomeric polyurethanes (TPUs) are the most common. Their structure is schematically shown in figure 3.1. The TPU consists basically of three building blocks: 1) a long-chain diol, normally with a polyether or polyester backbone, 2) a diisocyanate, mostly an aromatic one, 3) a chain extender, such as water, a short-chain diol, or a diamine. They are often prepared in a one pot procedure, in which the long-chain diol is first reacted with an excess of the diisocyanate, to form an isocyanate functionalized prepolymer. The latter is subsequently reacted with the chain extender which results in the formation of the high molecular weight polyurethane. If a diamine is used as the chain extender, the TPU also contains urea groups; however, it is still often called a polyurethane. At room temperature, the low melting soft blocks are incompatible with the high melting hard blocks, which induces microphase separation by crystallization or liquidliquid demixing. A schematic representation of the morphology of segmented copolymers is depicted in figure 3.2.5
40
Figure 3.2: Schematic representation of the morphology of a segmented block copolymer: () hard block, (~~~) soft block. The synthetic procedure to prepare TPUs has the intrinsic disadvantage that it leads to a distribution in the hard block lengths.6 As a result, the phase separation of these block copolymers is incomplete. Part of the hard blocks, in particular the shorter ones, are dissolved in the soft phase, causing an increase in the glass transition temperature, which is undesired for the low temperature flexibility of the material. The polydisperse hard block is manifested in a broad melting range and a rubbery plateau that is dependent on temperature. To improve these properties, and to get more insight into the structureproperties relationship, block copolymers containing hard blocks of uniform length have been prepared.7,8 These studies showed the advantages of uniform hard blocks, the polymers possessed better properties, which were less dependent on temperature. Several types of hard blocks were used to reach this objective, such as non-hydrogen bonding polyurethanes,7a,b normal polyurethanes,7c,d,f and polyurethane ureas.7e Niesten et al. used hard blocks based on uniform aramid units.8 Most TPEs containing uniform hard blocks were prepared by fractionation of a mixture of hard block oligomers, and subsequent copolymerization of the uniform hard oligomer of a specific length with the prepolymer. This procedure is hampered by the low solubility of especially the longer oligomers, and the difficulty in fractionating oligomers of different lengths, but very similar in chemical structure. In our opinion, the synthesis of uniform hard blocks via a protective group strategy would be advantageous, since this will open the way to exactly-defined hard blocks, presumably containing different monomer units, if desirable. Our choice was to prepare hard blocks based on urea units, making use of our knowledge of isocyanate chemistry. Urea groups are formed by a fast and complete reaction of an isocyanate with an amine. In the previous chapter, we showed that isocyanates were readily formed from amines by reaction with di-tert-butyl tricarbonate. This method is quite robust, and compatible with a variety of functional groups.9 41
Chapter 3
3.1.2 Urea-based TPEs Urea groups are known to associate strongly with one another via bifurcated hydrogen bonds (figure 3.3).10 In this way, the hydrogen bond strength exceeds that of amides and urethanes. The strong association between urea groups has been utilized by many researchers to obtain gelling agents.11 Both apolar and polar solvents, even water, were gelled by these compounds, as a result of intertwining of fibers composed of stacks containing hydrogen bonded urea groups.
N O N H H O N H N H O N H N H
Figure 3.3: Bifurcated hydrogen bonds between two urea groups. Thermoplastic elastomers containing urea groups have been prepared before.12 Only a few examples are known of TPEs possessing hard blocks comprising solely urea groups.13 Yilgr et al. described segmented polydimethylsiloxane and poly(ethylene oxide) based urea copolymers, and studied their behaviour in detail. These polymers were prepared by chain extension of amine-terminated prepolymers with several diisocyanates. However, this synthetic strategy does not allow the preparation of uniform hard blocks possessing more than two urea groups. The siloxane-based polymers possessed highly elastomeric properties, in contrast to the PEO-based ones. Interactions between urea groups and ether groups, in the latter polymer, lead to phase mixing and thus to poor mechanical properties. Interactions between urea groups and other functionalities, e.g. ether- or ester-groups, were also proposed with quantum mechanical calculations.13c An important drawback of ureas is their limited thermal stability. For ureas carrying two alkyl substituents, the decomposition temperature is about 200C.14 Substituting the alkyl groups by aromatic groups, lowers this temperature to below 160C. The presence of a catalyst, humidity, and disorder of the urea groups, also lower the decomposition temperature. In this chapter, the synthesis of segmented block copolymers possessing uniform (monodisperse) hard blocks based on urea groups is presented. The urea groups are expected to associate by hydrogen bonding (figure 3.4). The novel isocyanate chemistry described in chapter 2 is employed as a tool to prepare these materials in a controlled fashion.
42
Figure 3.4: Schematic representation of a block copolymer having a urea hard segment.
3.2 Molecular design of block copoly(ether urea)s Generally, the soft segments in thermoplastic elastomeric polyurethanes are made from long-chain diols, i.e. prepolymers with two telechelic hydroxy groups.1 This soft segment largely controls the low temperature properties. To obtain a material with high flexibility at room temperature, the soft phase should be completely amorphous, and its glass transition temperature should be as low as possible. The polarity of the soft phase plays an important role in the solubility and solvent resistance of the material. In addition to that, the strength of the hydrogen bonds in the hard block depends on the surrounding polarity, as well. A wide variety of soft segments with different polarities is available, ranging from the apolar polybutadiene glycols, to the water soluble poly(ethylene oxide). In this study, we mainly focused on poly(tetrahydrofuran) (1) also known as pTHF, poly(tetramethylene oxide), pTMO, poly(oxotetramethylene), Terathane as the soft segment.
HO 1 O H n
There are several reasons for this choice. pTHF is readily commercially available in a variety of molecular weights, and it is easily prepared on lab scale via a cationic ring-opening polymerization of the cheap monomer tetrahydrofuran. Further more, as we will show in chapter 6, it is possible to prepare monodisperse oligomers by a stepwise synthesis. The chemical and hydrolytic resistance of pTHF is high, due to its polyether main chain. Its glass transition temperature (86 to 70C)1,15 is far below room temperature, ensuring a high flexibility of the soft block at ambient temperature. pTHF of molecular weight higher than approximately 1500 g/mol is semicrystalline, with a melting point slightly above room 43
Chapter 3
temperature (32 to 43C). The crystallinity can be reduced by the copolymerization of THF with other monomers, e.g. ethylene oxide. pTHF is very soluble in a variety of organic solvents. Although it is moderately polar, it is not water-soluble, its low hygroscopicity is definitely an advantage compared to poly(ethylene oxide). pTHF is well-defined: its functionality is exactly 2.00, and hardly any cyclic oligomers are formed during its synthesis. For our purpose, the synthesis of block copoly(ether urea)s, prepolymers containing amine end groups are highly desired. Via functional initiation and termination of the polymerization of THF, various end groups can be introduced in pTHF.16 Amine end-capped pTHF has been prepared by termination of the polymerization with an excess of ammonia.17 However, besides primary amine groups, also secondary and tertiary amine groups are incorporated with this method. Another route is to obtain amine-terminated pTHF by an end group modification. One method, which has been previously used in our laboratory is cyanoethylation18 of hydroxy end groups, followed by hydrogenation.19
3.3 Synthesis of amine-terminated prepolymers Amine-terminated pTHF was synthesized by the above mentioned end group modification. Cyanoethylation is a Michael addition of acrylonitrile (scheme 3.1). A base catalyzes this reaction, since the nucleophilicity of a neutral alcohol is too low. Either potassium hydroxide or sodium hydride together with one equivalent of 15-crown-5 were used as a complexed base, in a 1% amount.
HO 1 O H n CN + base NC O 2 O CN n
Scheme 3.1: Cyanoethylation of pTHF. Using the latter base, the reaction is much faster, however, it is also more difficult to control. Since this reaction is exothermic and an equilibrium, it was performed at 0C, with a large excess of acrylonitrile to shift the equilibrium to the product side; in fact, acrylonitrile was used as the solvent. The anionic polymerization of acrylonitrile is a side-reaction that occurs during the cyanoethylation, resulting in the precipitation of a yellow solid. In general, as soon as the reaction mixture turned slightly yellow, the cyanoethylation was complete, as was most clearly evidenced in 13C-NMR by complete disappearance of the peak corresponding to the methylene adjacent to the alcohol group at 62.5 ppm. After quenching of the reaction by
44
addition of a small amount of hydrochloric acid, the reaction mixture was filtered or centrifuged, and concentrated. The nitrile groups of 2 were subsequently hydrogenated in order to obtain the amineterminated prepolymer 3 (scheme 3.2). We tried several ways to perform this reaction.
O 2 CN n BH 3 THF H 2N O 3
NC
NH2
Scheme 3.2. First, a catalytic hydrogenation using Raney/Cobalt was investigated. Different reaction conditions were employed. Toluene or methanol were used as solvent, ammonia was added to prevent side-reactions (imine and enamine formation), the amount of catalyst was varied, as well as reaction temperature and time. However, none of these reaction conditions gave conversions exceeding 90%. Prolonged reaction times resulted in a significant amount of retro-Michael addition. Furthermore, due to complexation of cobalt to the amines and/or ether functionalities in the product, it was difficult to remove the catalyst entirely from the product. This also hampered the characterization of the product by 1H-NMR. In conclusion, the catalytic hydrogenation of nitrile functionalized pTHF was not successful enough. Reduction of the nitrile groups with lithium aluminum hydride was very fast, but the strongly basic reaction conditions caused about 30% of retro-Michael addition, which is unsatisfactory. Finally, borane was employed to reduce the nitrile groups using a procedure reported by Brown.20 Reaction of prepolymer 2 with borane in THF at reflux temperature resulted in complete conversion within 4 hours.
b e f
H2N
d
O
e
O
NH2
4.5
3.5
2.5
2 (ppm)
1.5
0.5
-0.5
Figure 3.5: 1H-NMR spectrum of bis(3-aminopropyl)-pTHF2000 3. 45
Chapter 3
Workup of the reaction mixture by adding methanol and concentrated hydrochloric acid to destroy the boron-nitrogen complexes, followed by a basic extraction gave the amineterminated pTHF as a white powder. A small amount of butylated hydroxytoluene (0.1 w%) was added to stabilize the polymer from degradation by UV-light. Characterization of the product by 1H-NMR (figure 3.5), 13C-NMR, FT-IR, SEC and MALDI-TOF MS proved the proposed structure of 3 and showed a degree of functionalization of more than 97%. Table 3.1 shows the different amine-terminated polymers that were synthesized. The first two polymers are commercially available, and also contain 3-aminopropyl end groups. The others were synthesized according to the previously described procedure, except for the amine-terminated PEB (poly(ethylene-ran-butylene), hydrogenated polybutadiene), which was synthesized according to Hirschberg.19c pTHF4000 is a random copolymer of tetrahydrofuran and ethylene oxide, containing 10 mol% of the latter monomeric unit to reduce the crystallinity of the polymer. The hydroxy-terminated pTHF4000 was kindly provided by Akzo-Nobel. During synthesis of the amine-terminated polymer, some fractionation of the polymer occurred, resulting in an additional increase of the molecular weight of the product. In general, this effect is rather limited, only for the pTHF2900, a large difference in the molecular weight before and after reaction in observed. This is probably caused by an inadequate workup of this product. Table 3.1: Amine-terminated prepolymers. Prepolymer pTHF350 pTHF1100c pTHF2000 pTHF2900 pTHF4000d PEG2000 PEB3500
c
Backbone structure
O O O O n n n n
Mn,before (g mol1)a
Mn,after (g mol ) 350 1100 2500 5200 4500 2300 3600

1 b
Mw/Mn 1.8 1.7 1.5 1.5 1.4 n.d. n.d.
Yield (%)
O O
O x n
x y
2000 2900 4000 2000 3500
86 90 89 82 75
a) Molecular weight of original hydroxy-terminated prepolymer, determined by 1H-NMR. b) Molecular weight of amine-terminated prepolymer, determined by 1H-NMR. c) Commercially available. d) Random copolymer of tetrahydrofuran and ethylene oxide.
All the amine-terminated prepolymers presented in table 3.1 were used for the synthesis of thermoplastic elastomeric polyurethanes. We will focus mainly on the pTHF derived materials in the remainder of this chapter.
46
3.4 Synthesis of heterofunctional pTHF In the previous section, the synthesis of amine-terminated prepolymers was described. These telechelic polymers contain two amine end groups. From a macromolecular point of view, heterofunctional pTHF i.e. pTHF with two different end groups is very useful as a building block for more complex asymmetric architectures. Unfortunately, the number of pTHFs possessing two different end groups is limited. Some attempts were described in literature to obtain heterofunctional pTHF, e.g. -amino--carboxy-pTHF.21 But until now, no procedure for the synthesis of -amino--hydroxy-pTHF has been described. We aimed at obtaining these polymers by a functional initiation and a functional termination approach. This method was previously applied by Goethals et al. to prepare pTHFs with a variety of end groups.22 The first step involves the initiation of the cationic ring-opening polymerization of THF by a functional initiator. At the desired degree of polymerization or conversion, the living polymerization is terminated by addition of a functional terminator. In the last step, if necessary, both the initiator- and the terminator-group are deprotected. To obtain the -amino--hydroxy-pTHF, trimethylsilyl trifluoromethanesulfonate (TMS-OTf, 4), was used as functional initiator, and hexamethylenetetraamine (HMTA, 5) was used as functional terminator of the living polymerization of THF (scheme 3.3). TMSOTf results in a telechelic trimethylsilyl ether group, that is easily cleaved to restore a hydroxy group,23 and the HMTA derived end group is easily converted into a primary amine group.24
O Si O S CF3 + 4 Si O O O
n-2
O n
DTBP
N O+ + N N 5 N
N Si O O
n-1
MeOH / H2SO4 Toluene, reflux
HO 6
O
n-1
NH2
Scheme 3.3: Synthesis of -amino--hydroxy-pTHF 6. TMS-OTf has been reported previously as the initiator for the polymerization of THF. The main drawback of this initiator is its slow rate of initiation, due to both steric and
25
47
Chapter 3
electronic effects. This gives difficulties in controlling the molecular weight and polydispersity of the polymer. High molecular weights are already reached at low conversions of both THF and initiator. Since the molecular weights we are interested in ranges from 1000 2000 g/mol, we have to circumvent this problem. Therefore, the polymerization was carried out at reflux temperature (66C), to increase the initiation rate and to decrease the propagation rate. The rationale behind the decrease in propagation rate is the significant amount of depolymerization that occurs at reflux temperature, since it is close to the ceiling temperature of the bulk-polymerization of pTHF (Tc=83C).15,26 This results in higher polydispersities of the product (Mw/Mn1.52), whereas polydispersities slightly higher than 1 are expected for living polymerizations with fast initiation and no depolymerization. Prior to the addition of the initiator, a small amount of 2,6-di-tert-butylpyridine (DTBP) is added. This sterically hindered base acts as a non-nucleophilic proton trap and in this manner prevents initiation by trifluoromethane sulfonic acid, presumably formed by hydrolysis of TMS-OTf. The polymerization was terminated after 30 minutes by addition of HMTA, to form a quaternary ammonium end group. Subsequently, both end groups were deprotected in one step by heating the product in methanol and toluene containing some sulfuric acid. The polymers were purified by basic extraction. After workup, the product was successfully characterized by 1H-NMR (figure 3.6), 13C-NMR, FT-IR, MALDI TOF-MS, and SEC.
HO
O
n-2
b
NH2
BHT a b
BHT
4.5
3.5
2.5
2 (ppm)
1.5
0.5
-0.5
Figure 3.6: 1H-NMR spectrum of heterofunctional pTHF 6, Mn=1500 g/mol, BHT: butylated hydroxytoluene (stabilizer in diethyl ether).
48
Under normal conditions, the cationic ring-opening polymerization of THF is a living polymerization, in which it is easy to control the molecular weight (distribution). However, due to slow initiation and considerable amount of depolymerization in our case, it is quite difficult to control the molecular weight of this polymerization. Some control is achieved by varying the ratio between initiator and monomer, as is known for chain growth polymerizations.27 Figure 3.6 shows the molecular weights of the polymers as a function of the ratio between initiator and monomer. The molecular weights in figure 3.7 (right) were determined by 1H-NMR, and the polydispersities by SEC. Although some control of the molecular weights is achieved, the molecular weight distribution is rather broad. Figure 3.7 also shows that molecular weights of 1000 2000 g/mol, the range we are interested in, are not easily accessible, since quite high concentrations of initiator have to be used. Another drawback is the low conversion of both monomer and initiator under these conditions.
[I]/[M]
AU
time (min)
8000
6000 Mn (g/mol)
2.5 Mw/Mn
4000
2000
1.5
0 0 0.02 0.04 0.06 [I]/[M] 0.08 0.1
Figure 3.7: (top) SEC traces of 6 prepared with different initiator/monomer ratios; (bottom) Molecular weight as a function of initiator/monomer ratio. 49
Chapter 3
3.5 Preparation of block copoly(ether urea)s In the previous section, we described our aim to prepare thermoplastic elastomers in which the reversible crosslinking will merely be caused by hydrogen bonding between urea groups in the uniform hard blocks of the segmented copolymer. To obtain such well-defined polymeric structures, we aimed at using our novel isocyanate chemistry, in combination with employing protective group chemistry. In this section, the synthesis of segmented copolymers having an exact number of urea groups in the hard block is described. These polymers are schematically denoted as pTHFy-Ux, with x as the number of urea groups in the hard block, and y the number-averaged molecular weight of the pTHF soft block.
O O O m N H N H x n
pTHFy-Ux 3.5.1 Segmented copolymers with one urea group in the hard block The synthesis of a polymer with exactly one urea group in the hard block is shown in scheme 3.4. An amine-terminated prepolymer, pTHF with molecular weight of 1100 g/mol in this case, was reacted with one equivalent of di-tert-butyl tricarbonate 7 per amine group. The intermediate, with on average one isocyanate group, reacted immediately with another amine group resulting in a single urea group. After work-up, the product was obtained as a viscous liquid in a yield of 97%, and with Mw=68*103 g/mol based on SEC. The product is denoted as pTHF1100-U, in which U stands for a urea group.
O NH 2 pTHF NH2 + O O 7 O NH2 pTHF NCO 20C CHCl3 pTHF N H N H n O O O O 20C CHCl3
Scheme 3.4: Synthesis of pTHF1100-U. 3.5.2 Segmented copolymers with two urea groups in the hard block A polymer having a hard block comprising exactly two urea groups, was readily synthesized by chain extension of an amine-terminated prepolymer with a diisocyanate, or vice versa, an isocyanate-terminated prepolymer with a diamine (scheme 3.5). Which method 50
is preferred, depends on the availability of the appropriate starting compounds. The chain extender was added dropwise to a solution of the prepolymer, to ensure the exact 1 to 1 ratio of both functional groups in the equivalence point of the reaction, thus obtaining a maximum degree of polymerization. The extent of the reaction is easily followed by monitoring the isocyanate functionality by infrared spectroscopy. Due to the increase of the molecular weight of the polymer, and the association between urea groups, the solution became more viscous.
H2N pTHF NH2 +
O
OCN R NCO A
O O O O O
O pTHF N H N R N H H
O N H n
2 eq.
7 OCN pTHF NCO + H2N R NH2 B
Scheme 3.5: Two routes towards pTHF1100-U-R-U. Several types of chain extenders were used, from aliphatic to aromatic ones. Table 3.2 shows the chain extenders that were reacted with prepolymer pTHF1100. Most reactions were carried out in chloroform at room temperature. However, due to the low reactivity of aromatic diamines, these reactions were performed in pyridine at 90C. The polymers were obtained in good yields, and molecular weights were high (Mw=20 55*103 g/mol). All polymers were obtained as highly elastic fluffy fibers. Solution-casting of the polymers from chloroform solution gave completely transparent elastic films upon slow evaporation of the solvent. So, compared to the polymer with one urea group, incorporation of an additional urea group in the hard block leads to a remarkable improvement of the mechanical properties of the material, as was noticed by their physical appearance. Table 3.2: Chain extenders used for synthesis of pTHF1100-U-R-U. spacer method solventa T yield R (C) (%) n-C2H4 B chl 20 80 n-C3H6 B chl 20 86 A chl 20 89 n-C4H8 B chl 20 84 n-C5H10 A chl 20 89 n-C6H12 p-phenylene B pyr 90 85 m-phenylene A chl 20 70 o-phenylene B pyr 90 88 1,5-naphtylene B pyr 90 74 A chl 20 93 B pyr 90 66
CH 2
Mw (103 g/mol)b 41 47 42 32 43 27 38 50 20 55 28
a) chl: chloroform; pyr: pyridine. b) Measured by SEC, with NMP as solvent, relative to polystyrene standards.
51
Chapter 3
In order to study the influence of the nature and molecular weight of the prepolymer (soft block), several amine-terminated prepolymers (for characteristics see table 3.1) were chain extended by reaction with 1,4-diisocyanatobutane. These polymers are schematically denoted as e.g. pTHF1100-U2. For more synthetic details, see table 3.3. During the chain extension of the apolar PEB prepolymer, the solution turned into a gel. A small amount of HFIP was added to break the gel. After precipitation of the polymers, they were obtained as white, elastic, fluffy fibers, except for pTHF350-U2, which was obtained as a powder. The molecular weight of this polymer was also considerable lower than that of the other polymers, probably caused by the low functionality of the starting amine-terminated prepolymer. Table 3.3: Polymers possessing hard blocks comprising two urea groups, separated by a butylene spacer. prepolymer pTHF350 pTHF1100 pTHF2000 pTHF2900 pTHF4000 PEG2000 PEB3500 method A A A A A A A solvent chl chl chl chl chl chl chl + HFIP T (C) 20 20 20 20 20 20 20 yield (%) 83 89 85 84 93 Mw (103 g/mol)a 8.5 42 53 32 58 n.d. n.d.
a) chl: chloroform; pyr: pyridine; HFIP: 1,1,1,3,3,3-hexafluoro-isopropanol. b) Measured by SEC, with NMP as solvent, relative to polystyrene standards.
3.5.3 Segmented copolymers with three urea groups in the hard block Although the copoly(ether urea)s with two urea groups in the hard block possess already highly elastic properties, we also aimed at synthesizing a polymer comprising three urea groups in the hard block, pTHF1100-U3, since it will have a higher melting temperature of the hard block. The synthetic strategy to achieve this goal is outlined in scheme 3.6. The first step was the mono-protection of 1,4-diaminobutane by reaction with di-tert-butyl dicarbonate. After several extractions and a recrystallization 4-(tert-butoxycarbonylamino)-1-butylamine 8 was obtained in a yield of 80%. Reaction of the remaining amine group with di-tert-butyl tricarbonate gave isocyanate 9 within 15 min in a quantitative yield. Isocyanate 9 is an important building block for the synthesis of uniform oligourea blocks. The corresponding compounds with an ethylene and propylene spacer could not be prepared, since they were not stable. An intramolecular reaction between the isocyanate functionality and the NH of the carbamate group resulted in spontaneous ring closure into five-, or six-membered rings, respectively . 52
Isocyanate 9 was reacted with the amine-terminated pTHF1100, and in the next step the protective group was removed by treatment with trifluoroacetic acid (TFA), yielding the amine-terminated prepolymer with one urea group at both end groups 11. Reaction of 11 with two equivalents of di-tert-butyl tricarbonate 7 gave the corresponding isocyanate-terminated prepolymer, and polycondensation with amine-terminated prepolymer 11 resulted in the segmented copolymer with exactly three urea groups in the hard block. During this step some methanol was added to avoid gelation. After precipitation of the product in pentane, the polymer was obtained as a white fluffy powder. Solution-casting of the product gave a transparent flexible film. The molecular weight could not be determined by SEC, since the polymer was insoluble in the eluent, NMP.
O O O O CH2Cl2 O O CHCl3 O H2N pTHF NH2 + 2 O N H H O O O H2N N H O OCN N H O pTHF N H N H N H H N O pTHF1100-U3 pTHF N H H N N H N H pTHF 11 N H O N H O N H n NCO + 11 MeOH/CHCl3 N N H O N H pTHF 10 O N H NH2 + 2 7 CHCl3 N H O N H NCO CHCl3 O N H 9 NCO O N H 8 NH2 H2N excess O 8 + O O 7 O O O NH2 + O O
H N O O
TFA CH2Cl2
Scheme 3.6.
53
Chapter 3
3.5.4 Segmented copolymers with four urea groups in the hard block The synthesis of a segmented copoly(ether urea) having a hard block containing exactly four urea groups in the hard block, pTHF1100-U4, is a straightforward continuation of the synthesis of the one with three urea groups. Amine-terminated prepolymer 11 was reacted with 1,4-diisocyanatobutane in a mixture of methanol and chloroform (scheme 3.7). Due to very strong interaction between this hard block comprising four urea groups, the solution became highly viscous, and this hampered the precipitation of the product in heptane. The polymer was obtained as a rubbery solid. The molecular weight could not be determined by SEC, since the polymer was insoluble in the eluent, NMP.
O O H2N N H N H pTHF N 11 H N H N O pTHF1100-U4 H N H NH2 + OCN NCO MeOH CHCl3
O pTHF N H
H N N H
O N H
H N O
H N n
Scheme 3.7. 3.5.5 Segmented copolymers with a polydisperse hard block All segmented copoly(ether urea)s described above possess uniform hard blocks comprising an exact number of urea groups in the hard block. To study the influence of the uniformity of the hard block, a segmented copoly(ether urea) was prepared having a polydisperse hard block with on average two urea groups per hard block, pTHF1100-U2PD. This polymer was prepared by mixing bis-terminated pTHF1100 and 1,4-diaminobutane stoichiometrically, and by subsequently adding half an equivalent of di-tert-butyl tricarbonate 7 per amine group (scheme 3.8). Half of the amine groups reacted with 7, and were converted to isocyanate groups. These reacted with the other half of amine groups, yielding hard blocks with a distribution of the number of urea groups, but on average two. Equal reactivity of the amine groups is assumed, which will result in a random distribution of the hard block length. After precipitation of the polymer, it was obtained as white, elastic, fluffy fibers. It is moderately soluble in chloroformmethanol mixtures, but it is insoluble in NMP, hampering the determination of its molecular weight. No evidence of complete randomness of the hard block length is present, but it is assumed based on statistics.
54
O H2N pTHF NH2 + H2N NH2 O pTHF N H N H 2 n + 2 O O
O O 7
O O
pTHF1100-U2PD
Scheme 3.8: Synthesis of polymer with polydisperse hard block, pTHF1100-U2PD.
3.6 Hydrogen bonding within block copoly(ether urea)s The synthesis of block copoly(ether urea)s with uniform hard blocks was described in the previous section. The number of urea groups in the hard blocks ranged from 1 to 4. All polymers were characterized by 1H-NMR and FT-IR. These spectra were in agreement with the proposed structures. The molecular weights of the polymers that were soluble in NMP were determined by SEC, and shown in table 3.2 and 3.3. Infrared spectroscopy is a helpful tool for studying the extent of hydrogen bonding of several hydrogen bonding groups.13f,28 Coleman and Painter reported on an temperature-dependent infrared study of a polyurea and model ureas.28a They concluded that the frequency of both the NH and the C=O (amide I) vibration depend strongly on the hydrogen bonding nature of the urea groups (figure 3.8).
=1615 cm-1 (strong)
1640 cm-1 (weak)
=3400 cm-1
N H
O N H
=1690 cm
-1
O N H O N N H N H
=3325 cm-1 (strong)

3350 cm-1 (weak)
'Free' urea group
Hydrogen bonded urea group
Figure 3.8: Characteristic infrared bands for urea groups. If the hydrogen bonding strength between urea groups increases, the frequency of both the N H and C=O vibrations decreases, in contrast to the amide II band (a combined NH bending and CN stretching vibration) at approximately 1575 cm-1, that increases in frequency. Thus, the position of these bands is a direct indication of the strength of hydrogen bonding. Furthermore, competitive hydrogen bonding between urea and ether groups in the soft block, 55
Chapter 3
so called mixing of hard and soft phases, was quantified. It is more practical to use the carbonyl stretching vibration, since this band is more narrow than the NH stretching vibration, and it shifts more upon increase of the hydrogen bonding strength. In our copolymers, the absence of any carbonyl groups other than from the urea groups, such as urethanes, amides or esters is advantageous, allowing an unambiguous analysis of the hydrogen bonds. The block copoly(ether urea)s show a remarkable difference in their physical properties. The polymer with only one urea group in the hard block, pTHF1100-U, possesses hardly any mechanical properties; it was obtained as a viscous liquid. This is an indication of the absence of stable crosslinks at room temperature. In the infrared spectra this is evidenced by the position of the carbonyl stretching vibration at 1637 cm-1 (figure 3.9 top), corresponding to weakly hydrogen bonded urea groups. Apparently, one single urea group is not sufficient to form stable crosslinks by strong hydrogen bonds. Hence, this unit cannot form a hard block. However, for the sake of conformity, we like to use this term here.
pTHF1100-U
1637
1568 pTHF1100-U2
%T
1615
1580
pTHF1100-U3
1615
1576
pTHF1100-U4
1616
1574
1700
1675
1650
1625
1600 -1 (cm )
1575
1550
1525
1500
Figure 3.9: Carbonyl region of infrared spectra of copoly(ether urea)s. The segmented copolymers with two urea groups in the hard block (table 3.2 and 3.3) possess very intriguing elastic properties. The fibers are shape-persistent and not tacky. The infrared spectrum of pTHF1100-U2 (figure 3.9) shows a strong, sharp peak at 1615 cm-1, indicative of strong hydrogen bonding between urea groups. No other peaks corresponding to weakly hydrogen bonded and free urea groups were observed. Thus, two urea groups in the 56
hard blocks induces strong association and stable crosslinks at room temperature. In table 3.2 the different spacers between the urea groups are shown. These polymers possess comparable properties. In all polymers with two urea groups, strong hydrogen bonding is present according to infrared spectroscopy. The frequencies for the aromatic hard blocks are fairly higher than for the aliphatic ones, this is an effect of the substituents of the urea group, and not of the extent of hydrogen bonding. Table 3.3 shows the different soft blocks that were used in the segmented polymers with two urea groups in the hard block. Upon comparison of the FT-IR spectra of the polymers based on pTHF, no difference is observed for the position of the carbonyl stretching vibration. However, the intensity of this peak varies considerably, since the concentration of urea groups depends on the molecular weight of the soft block. In contrast to the polymers based on pTHF, the one based on PEG is sticky and IR shows that there are no strong hydrogen bonds present (carbonyl stretching vibration at 1646 cm-1). The NH region of the IR spectrum shows a broad peak indicative of different hydrogen bonding modes, from nonhydrogen bonded to weakly bonded NH groups. Competitive hydrogen bonding between urea groups and ether groups is present as well. This phase-mixing is often observed for block copolymers based on poly(ethylene oxide).13c,f,28a Thus, the polar PEG-matrix and the abundance of ether groups hampers the formation of stable crosslinks by strong hydrogen bonding between the hard blocks. The polymer based on the very apolar PEB soft block possesses properties comparable to pTHF-based ones. The bis-urea hard blocks associate strongly in this apolar matrix, which is also evidenced by infrared spectroscopy. The copoly(ether urea)s having three and four urea groups in the hard block show comparable FT-IR spectra as the pTHF based polymers with two urea groups (figure 3.9). The intensity of the bands corresponding to urea groups increases with increasing number of urea groups. The position of the amide I band does not change upon introducing more than two urea groups, indicating no change in hydrogen bonding strength. The amide II band slightly shifts and broadens, the reason for this is not known. To compare the hydrogen bonding in the uniform hard blocks with the polydisperse ones, part of the infrared spectra of polymers pTHF1100-U2 and pTHF1100-U2PD is shown in figure 3.10. Both polymers have the same overall composition, which is reflected by the equal intensity of the different functional groups in the polymer. However, the polymer with the polydisperse hard block shows a broader peak corresponding to the carbonyl of the urea group compared to the uniform hard block. This indicates a distribution in hydrogen bonding strengths within the polydisperse hard blocks. The position of this peak is also at higher frequency (1620 cm-1 versus 1615 cm-1, respectively), which denotes that, on average, the hydrogen bonding in the polydisperse hard blocks is weaker than in the uniform hard blocks. 57
Chapter 3
This confirms our hypothesis that uniform hard blocks associate stronger than polydisperse hard blocks. The effect thereof on the material properties is studied in the next chapter.
pTHF1100-U2
%T
1615
1580
pTHF1100-U2PD
1620 1579
1700
1675
1650
1625
1600 -1 (cm )
1575
1550
1525
1500
Figure 3.10: Carbonyl region of infrared spectra of copoly(ether urea)s: uniform hard block (top), and polydisperse hard block (bottom).
3.7 Conclusions Block copoly(ether urea)s possessing uniform hard blocks were prepared. Amineterminated prepolymers are one of the building blocks of these segmented copolymers. They were prepared from commercially available hydroxy-terminated prepolymers via a two-step synthesis. The first step involved a cyanoethylation (Michael addition of acrylonitrile) and subsequently the nitrile end groups were hydrogenated by reaction with borane. The ,diamino-poly(tetrahydrofuran)s were prepared on a 50 gram scale and obtained in excellent yields. -Amino--hydroxy-poly(tetrahydrofuran), an asymmetric prepolymer having two different end groups, was prepared by cationic ring opening polymerization of tetrahydrofuran. Employing a strategy of functional initiation and functional termination, both end groups were introduced selectively. However, due to relatively slow initiation, the molecular weight was difficult to control and polydispersities are rather high. The amine-terminated prepolymers were used as the soft block in the block copoly(ether urea)s. Via protective group strategy and novel isocyanate chemistry, we 58
prepared uniform hard blocks consisting of exactly 1 to 4 urea groups. The strength of hydrogen bonding of the urea groups in the hard block was studied by infrared spectroscopy. The polymer having only one urea group in the hard block is a viscous liquid, and FT-IR shows the absence of strong hydrogen bonds between the urea groups in the hard block. Polymers having two urea groups in the hard block possess very appealing properties, being highly elastic and very soluble. Their synthesis is very attractive as well, because no elaborate protective group chemistry has to be employed. Strong hydrogen bonding is observed by FTIR. A number of polymers with varying spacers between the urea groups, and with different soft blocks were prepared. At first glance, the spacer had little influence on the properties. The polarity of the soft block had a pronounced effect; in the polar poly(ethylene oxide)based soft block, the hydrogen bonds are weaker than in the less polar poly(tetrahydrofuran) and apolar poly(ethylene-ran-butylene)-based soft blocks. Increasing the number of urea groups in the hard block even further, gave insoluble gel-forming polymers, hard to process. Apparently, hydrogen bonding between these hard blocks is too strong. Comparison of the infrared spectra of a polymer possessing uniform hard blocks, with one possessing a polydisperse hard block, indicates a distribution of hydrogen bonding strengths and, on average, weaker hydrogen bonding within the polydisperse hard block. In the next chapter the morphology and properties of the block copoly(ether urea)s will be described in more detail. 3.8 Experimental section
General methods and instrumentation. General methods concerning purification of solvents, spectroscopic techniques and mass spectrometric techniques can be found in the experimental section of chapter 2. MALDI-TOF spectra were obtained at a Perseptive Biosystems Voyager DE-Pro MALDI-TOF mass spectrometer (accelerating voltage: 20kV; grid voltage: 74.0%, guide wire voltage: 0.030%, delay: 200 ms, low mass gate 900 amu). Samples for MALDI-TOF were prepared by adding a solution of the polymers in THF (20 l, c=1 mg/ml) to a solution of -cyano-4hydroxycinnamic acid in THF (10 l, c=20 mg/ml) and subsequent thoroughly mixing. This mixture (0.3 l) was brought on a sample plate, and the solvent was evaporated. Size exclusion chromatography (SEC) was performed on a Shimadzu LC10-AT, using a Polymer Laboratories Plgel 500 A column, a Shimadzu SPD-10AV UV-Vis detector, and chloroform as eluent. Polystyrene standards were used for calibration. Materials. Bis(3-aminopropyl)-poly(tetrahydrofuran) with molecular weights 350 and 1100 g/mol, poly(tetrahydrofuran) with molecular weights 2000 and 2900 g/mol, and poly(ethylene oxide) with molecular weight 2000 were purchased from Aldrich. Random copolymer of THF and EO of molecular weight 4000 g/mol was kindly provided by prof. dr. Doetze Sikkema (Akzo-Nobel), and hydroxy-terminated poly(ethylene-ranbutylene) (hydrogenated polybutadiene, Kraton liquid polymer L-2203) was kindly provided by Kraton Polymers Research BV. 1,4-Diisocyanatobutane, 1,3-phenylenediisocyanate, 1,5-naphthalenediamine, methylenedianiline, borane-tetrahydrofuran complex (1 M in THF), and sodium hydride (60% dispersion in mineral oil) were purchased from Aldrich; 1,3-diaminopropane and 1,2-ethanediol from Acros; 1,2ethylenediamine from Janssen; 1,5-diaminopentane, 1,6-diisocyanatohexane, 1,4-phenylenediamine, and 1,2-
59
Chapter 3
phenylenediamine from Fluka. 1,4-Bis(4-aminophenylethynyl)-benzene was kindly provided by Albert Schenning. Bis(2-cyanoethyl)-poly(tetrahydrofuran), 2 Poly(tetrahydrofuran) diol, Mn=2000 g/mol, (20.00 g; 10.0 mmol) and 15-crown-5 (44 mg; 0.2 mmol) were dissolved in acrylonitrile (40 ml) and cooled on an icebath. Sodium hydride (8 mg 60% dispersion in mineral oil; 0.2 mmol) is added to the solution, and the reaction mixture is stirred at 0C for about 15 min, after which the reaction mixture turned slightly yellow. At this point, the reaction was quenched by addition of a drop of concentrated hydrochloric acid. The solution was concentrated, the residue taken up in dichloromethane (100 ml) and centrifuged at 4500 rpm. The mixture was decanted, filtered, and concentrated in vacuo. The product was obtained as a slightly yellow, viscous liquid, that slowly crystallized (20.13 g, 96%). 1H-NMR (CDCl3): 3.62 (t, 4H, OCH2CH2CN), 3.51 (t, 4H, CH2OCH2CH2CN), 3.40 (br. t, 106H, OCH2CH2CH2CH2O main chain), 2.59 (t, 4H, CH2CN), 1.60 (br. m, 110H, OCH2CH2CH2CH2O main chain). 13C-NMR (CDCl3): 117.7 (CN), 71.0 (CH2OCH2CH2CN), 70.4 (OCH2CH2CH2CH2O main chain), 65.1 (OCH2CH2CN), 26.3 (OCH2CH2CH2CH2O main chain), 18.7 (CH2CN). FT-IR (ATR): 2939, 2855, 2161 (w, CN stretching), 1367, 1103 (CO stretching) cm-1. Bis(3-aminopropyl)-poly(tetrahydrofuran), 3 To a solution of borane-tetrahydrofuran complex (80 ml 1M in THF, 80 mmol) in dry THF (240 ml) was added slowly bis(2-cyanoethyl)-poly(tetrahydrofuran) 2 (20.00 g, 9.5 mmol) dissolved in dry THF (160 ml) at 0C. The solution was stirred for 30 min at 0C, after which it was heated to reflux for 4 h. The reaction mixture was cooled to 0C, and methanol (80 ml) was added dropwise. (Be careful: hydrogen-gas evolution). Hydrochloric acid (4 ml, 37% in water) was added slowly, and the reaction mixture was stirred for 1 h, and subsequently evaporated to dryness under reduced pressure. Trimethyl borate was removed by three coevaporations with methanol (3 times 100 ml). To the viscous liquid was added sodium hydroxide solution (150 ml, 1M in water), and this was extracted with diethyl ether (3 times 300 ml). The combined organic layers were dried with sodium sulfate, filtered, and the solvent was evaporated on a rotary evaporator without putting the flask in the waterbath. During the evaporation, the polymer precipitated from the cold solution and was obtained as a white powder (18.74 g, 93%). 1H-NMR (CDCl3): 3.49 (t, 4H, OCH2CH2CH2NH2), 3.41 (br. t, 138H, OCH2CH2CH2CH2O main chain), 2.79 (t, 4H, CH2NH2), 1.71 (t, 4H, OCH2CH2CH2NH2), 1.62 (br. m, 142H, OCH2CH2CH2CH2O main chain), 1.1 (br. s, 4H, NH2). 13C-NMR (CDCl3): 70.5 (OCH2CH2CH2CH2O main chain), 68.8 (OCH2CH2CH2NH2), 39.7 (CH2NH2), 33.6 (OCH2CH2CH2NH2), 26.4 (OCH2CH2CH2CH2O main chain). FT-IR (ATR): 3564, 3539, 2941, 2862, 1492, 1372, 1107, 996 cm-1. MALDI-TOF [M+Na+] = Calcd. 155.1 + n*72.0 Da. Obsd. 155.9 + n*72.0 Da. SEC (phenyl urea derivative): Mn=3769 g/mol, PDI=1.5. -(4-Aminobutyl)--hydro-poly(tetrahydrofuran), 6 Freshly distilled tetrahydrofuran (5.00 ml, 61.6 mmol) was inserted into a dried 50 ml flask via a syringe under exclusion of air, and 2,6-di-tert-butylpyridine (139 l, 0.618 mmol) was added to this. This solution was heated to reflux under an argon atmosphere. Trimethylsilyl trifluoromethanesulfonate (446 l, 2.47 mmol) was added, and the reaction mixture was stirred for 30 min at reflux under an argon atmosphere. The reaction was quenched by addition of hexamethylene tetraamine (0.866 g, 6.18 mmol) dissolved in chloroform (20 ml). The turbid reaction mixture was stirred for another 30 min, after it was evaporated to dryness under reduced pressure. The residue was redissolved in methanol (30 ml) and toluene (20 ml), and sulfuric acid (1.5 ml) was added. The reaction mixture was heated to 100C for 3 h, and after cooling down, it was neutralized by addition of a 1 M sodium hydroxide solution. After partially concentration of the reaction mixture, 1 M sodium hydroxide solution (20 ml) was added, and the reaction mixture was extracted with diethyl ether (two times 20 ml). The combined organic layers were concentrated, and the residue redissolved in THF (5 ml). The solution was precipitated in 1 M sodium hydroxide solution (50 ml) at 0C, and the white product was filtered off, washed with the sodium hydroxide solution, and redissolved in diethyl ether (20 ml). The solution was dried with sodium sulfate and evaporated to dryness under reduced pressure while cooling the flask on an ice bath, allowing a white powder to crystallize (0.67 g). 1H-NMR (CDCl3, 400 MHz): 3.64 (t, 2H, CH2OH), 3.45 (m, 384H, CH2OCH2), 2.71 (t,
60
2H, CH2NH2), 2.61 (s, 1H, OH), 1.65 (m, 392H, CH2CH2CH2), 1.04 (s, 2H, NH2) ppm. 13C-NMR (CDCl3, 100 MHz): 70.6 (CH2OCH2), 62.6 (CH2OH), 42.1 (CH2NH2), 30.8 (CH2CH2NH2), 30.3 (CH2CH2OH), 26.5 (CH2CH2CH2) ppm. Mn(1H-NMR)=7000 g/mol. MALDI-TOF [M+Na+] = Calcd. 184.1 + n*72.0 Da. Obsd. 184.8 + n*72.0 Da. pTHF1100-U Bis(3-aminopropyl)-poly(tetrahydrofuran), Mn=1100 g/mol, (2,00 g, 1.82 mmol) was dissolved in chloroform (20 ml), and a solution of di-tert-butyl tricarbonate (0.48 g, 1.82 mmol) in chloroform (4 ml) was injected into this solution. The reaction mixture was stirred for 2 h, and subsequently evaporated to dryness under reduced pressure. The product was obtained as a very viscous colourless oil. 1H-NMR (DMSO): 3.33 (60H, CH2O), 2.95 (4H, CH2N), 1.50 (65H, CH2CH2CH2). FT-IR (ATR): 3350 (NH stretching), 2938, 2853, 1637 (C=O stretching), 1568, 1366, 1103 (CO stretching) cm-1. SEC (NMP, rel. to PS): Mn=68*103 g/mol. pTHF1100-U-C2H4-U Via route A (scheme 3.5): Bis(3-aminopropyl)-poly(tetrahydrofuran), Mn=1100 g/mol, (0.50 g, 0.45 mmol) was dissolved in chloroform (10 ml), and a solution of di-tert-butyl tricarbonate (0.235 g, 0.91 mmol) in chloroform (1 ml) was injected into this solution. The reaction mixture was stirred for 30 min. 1,2-Ethylenediamine (0.0269 g, 0.45 mmol) in chloroform (3 ml) was added dropwise, and the solution was stirred for 1 h, and subsequently partly concentrated, and methanol (1 ml) was added. The product was precipitated in pentane (50 ml), filtered and dried in vacuo. It was obtained as white, fluffy, elastic fibers (0.47 g, 86%). 1H-NMR (DMSO): 5.91 (4H, NH), 3.34 (59H, CH2O), 2.99 (8H, CH2N), 1.51 (56H, CH2CH2CH2). FT-IR (ATR): 3329 (NH stretching), 2937, 2854, 1615 (C=O stretching), 1589 1366, 1105 (CO stretching) cm-1. SEC (NMP, rel. to PS): Mn=41*103 g/mol. pTHF1100-U-C4H8-U (= pTHF1100-U2) Via route B (scheme 3.5): Bis(3-aminopropyl)-poly(tetrahydrofuran), Mn=1100 g/mol, (10.00 g, 9.09 mmol) was dissolved in chloroform (100 ml), and a solution of 1,4-diisocyanatobutane (1.40 g, 9.99 mmol) in chloroform (40 ml). About 75% of the latter solution was added to the former solution at once, the rest was added dropwise. The solution was stirred for 1 h, and subsequently partly concentrated, and methanol (5 ml) was added. The product was precipitated in hexane (500 ml), filtered and dried in vacuo. It was obtained as white, fluffy, elastic fibers (10.62 g, 93%). 1HNMR (CHCl3): 5.44.8 (4H, NH), 3.41 (58H, CH2O), 3.25 (4H, OCH2CH2CH2N), 3.17 (4H, NCH2CH2CH2CH2N), 1.74 (4H, OCH2CH2CH2N), 1.62 (58H, OCH2CH2CH2CH2O), 1.50 (4H, NCH2CH2CH2CH2N). FT-IR (ATR): 3324 (NH stretching), 2940, 2854, 1615 (C=O stretching), 1580 1365, 1104 (CO stretching) cm-1. SEC (NMP, rel. to PS): Mn=42*103 g/mol. =0.98 g/cm3. 4-(Tert-butoxycarbonylamino)-1-butylamine, 8 Di-tert-butyl dicarbonate (14.40 g, 50 mmol) was dissolved in chloroform (100 ml), and added dropwise to a solution of 1,4-diaminobutane (17.60 g, 200 mmol) in chloroform (150 ml) at 0 C. The suspension was stirred overnight at 20C. The reaction mixture was washed three times with water (100 ml) and the product was extracted by hydrochloric acid solution (100 ml 1 M in water). The aqueous layer was isolated, basified by addition of sodium hydroxide solution (10 ml 10 M in water), and subsequently extracted 3 times with dichloromethane (3 times 100 ml). The combined organic layers were dried with sodium sulfate, filtered and the dichloromethane was removed under reduced pressure. Recrystallization of the product from diisopropyl ether gave the pure product as colourless crystals (7.80 g, 80%, with respect to di-tert-butyl dicarbonate). Tm=84C. 1 H-NMR (CDCl3): 4.71 (s, 1H, NHBoc), 3.13 (q, 2H, BocNH-CH2, J=6.2 Hz), 2.71 (t, 2H, H2NCH2, J=6.6 Hz), 1.57-1.46 (m, 4H, CH2CH2CH2), 1.44 (s, 9H, CH3), 1.28 (s, 2H, NH2) ppm. 13C-NMR (CDCl3): 155.8 (C=O), 79.0 ((CH3)3C), 41.9 (CH2NHBoc), 40.5 (CH2NH2), 31.0 (CH2CH2NHBoc), 28.5 (CH3), 27.6 (CH2CH2NH2) ppm. FT-IR (ATR): 3359 (NH stretching), 1689 (C=O stretching), 1521, 1167 cm-1. Anal. Calcd. (%) for C9H20N2O2 (188.27): C 57.42; H 10.71; N 14.88. Found (%): C 56.65; H 10.62; N 14.58.
61
Chapter 3
O,O-Bis(4-(tert-butoxycarbonylamino)butylureido)-poly(tetrahydrofuran), 10 Di-tert-butyl tricarbonate (1.03 g, 3.94 mmol) was dissolved in chloroform (20 ml), and 4-(tertbutoxycarbonylamino)-1-butylamine (0.74 g, 3.94 mmol) in chloroform (4 ml) was injected in this solution. The reaction mixture was stirred for 30 min, and bis(3-aminopropyl)-poly(tetrahydrofuran), Mn=1100 g/mol, (2.00 g, 1.82 mmol) in chloroform (8 ml) was added, and the solution was stirred for 1 h, after which it was concentrated in vacuo (2.75 g, 99%). 1H-NMR(CDCl3): 5.4 (br. s, 2H, NHBoc), 4.8 (br. s, 4H, NHC=ONH), 3.53 (t, 4H, OCH2CH2CH2N), 3.42 (br. m, 60H, OCH2CH2CH2CH2O), 3.31 (t, 4H, OCH2CH2CH2N), 3.16 (m, 8H, CH2CH2CH2CH2N), 1.80 (qui, 4H, OCH2CH2CH2N), 1.62 (br. m, 70H, OCH2CH2CH2CH2O), 1.50 (m, 4H, NCH2CH2CH2CH2N), 1.44 (s, 18H, CH3). FT-IR (ATR): 3357, 1680, 1626, 1579, 1520, 1365, 1102 cm-1. O,O-Bis(4-aminobutylureido)-poly(tetrahydrofuran), 11 O,O-Bis(4-(tert-butoxycarbonylamino)butylureido)-poly(tetrahydrofuran) 10 (6,80 g, 8.95 mmol) was dissolved in dichloromethane (10 ml) and trifluoroacetic acid (5 ml) was added. The reaction mixture was stirred at 20C overnight. Dichloromethane and trifluoroacetic acid were removed under reduced pressure and the residue was redissolved in dichloromethane (50 ml) and washed with sodium hydroxide solution (50 ml 1 M in water). The aqueous layer was extracted two more times with dichloromethane (two times 50 ml). The combined organic layers were dried with sodium sulfate, filtered and the dichloromethane was removed under reduced pressure. The product was obtained as a waxy solid (4.90 g, 85%). 1H-NMR (CDCl3): 4.9-4.7 (d br. s, 4H, NHC=ONH), 3.50 (t, 4H, NCH2CH2CH2O, J=5.9 Hz), 3.41 (br. m, 70H, OCH2CH2CH2CH2O), 3.27 (q, 4H, OCH2CH2CH2N, J=6.2 Hz), 3.27 (q, 4H, (C=O)NHCH2CH2CH2CH2N, J=5.9 Hz), 2.71 (t, 4H, NH2CH2, J=6.8 Hz), 1.75 (qui, 4H, NCH2CH2CH2O, J=6.3 Hz), 1.62 (br. m, 70H, OCH2CH2CH2CH2O), 1.50 (m, 8H, NHCH2CH2CH2CH2N) , 1.03 (s, 4H, NH2) ppm. FT-IR (ATR): 3327, 2938, 2859, 1616, 1587, 1365, 1106 cm-1. pTHF1100-U3 O,O-Bis(4-aminobutylureido)-poly(tetrahydrofuran) 11 (0.50 g, 0.368 mmol) was dissolved in chloroform (5 ml) and di-tert-butyl tricarbonate (0.19 g, 0.735 mmol) in chloroform (1 ml) was injected into this solution. The reaction mixture was stirred for 30 min, and methanol (3 ml) was added, immediately followed by O,O-bis(4aminobutylureido)-poly(tetrahydrofuran) 11 (0.50 g, 0.368 mmol) in chloroform (3 ml). The reaction mixture was stirred for 1 h, and subsequently precipitated in pentane (100 ml). The product was obtained as a white fluffy powder (0.89 g, 88%). 1H-NMR (10% TFA in CDCl3): 3.71 (64H, CH2O), 3.45 (4H, OCH2CH2CH2N), 3.32 (8H, NCH2CH2CH2CH2N), 1.99 (4H, OCH2CH2CH2N), 1.73 (71H, OCH2CH2CH2CH2O + NCH2CH2CH2CH2N). FT-IR (ATR): 3324, 2940, 2854, 1615, 1576 1366, 1105 cm-1. pTHF1100-U4 O,O-Bis(4-aminobutylureido)-poly(tetrahydrofuran) 11 (0.50 g, 0.368 mmol) was dissolved in chloroform (7 ml) and methanol (3 ml), 1,4-Diisocyanatobutane (0.0567 g, 0.405 mmol) was dissolved in chloroform and added dropwise to the former solution. Subsequently, the reaction mixture was stirred for 1 h. The viscous solution was poured into heptane and the suspension was stirred for 3 h. Then it was allowed to settle, the liquid was decanted, and the product was dried under reduced pressure. It was obtained as a rubbery solid (0.40 g, 73%). 1H-NMR (10% TFA in CDCl3): 3.71 (66H, CH2O), 3.45 (4H, OCH2CH2CH2N), 3.31 (12H, NCH2CH2CH2CH2N), 1.98 (4H, OCH2CH2CH2N), 1.72 (73H, OCH2CH2CH2CH2O + NCH2CH2CH2CH2N). FTIR (ATR): 3323, 2940, 2855, 1616, 1574 1368, 1104 cm-1. pTHF1100-U2PD Bis(3-aminopropyl)-poly(tetrahydrofuran), Mn=1100 g/mol, (2.00 g, 1.79 mmol), and 1,4-diaminobutane (0.158 g, 1.79 mmol) were dissolved in chloroform (20 ml). A solution of di-tert-buty tricarbonate (0.94 g, 3.58 mmol) in chloroform (3 ml) was injected into this solution. The reaction mixture was stirred for 1 h at room temperature, during which the solution formed a gel. Methanol (5 ml) was added and the solution was partly concentrated and then precipitated in pentane (150 ml). The product was obtained as white, fluffy, elastic fibers (2.02 g, 90%). 1H-NMR (CDCl3/methanol-d4): 3.34 (80H, CH2O), 3.21 (4H, ), 3.13 (4H, NCH2CH2CH2CH2N), 1.75 (4H, OCH2CH2CH2N), 1.64 (77H, OCH2CH2CH2CH2O), 1.49 (4H,
62
NCH2CH2CH2CH2N). FT-IR (ATR): 3326 (NH stretching), 2939, 2853, 1620 (C=O stretching), 1579 1366, 1104 (CO stretching) cm-1. SEC (NMP, rel. to PS): Mn=n.d.

1. Legge, N. R., Holden, G. and Schroeder, H. E. Thermoplastic elastomers: A comprehensive review; Carl Hansser Verslag: New York, 1987. 2. a) Coleman, D. J. Polym. Sci. 1954, 14, 15; b) Witsiepe, W. K. ACS Advances in Chemistry 1973, 129, 39. 3. Deleens, G., Foy, P. and Marechal, E. Eur. Pol. Journal 1977, 13, 337. 4. Bayer, O., Muller, E., Petersen, S., Piepenbrink, H. F. and Windemuth, E. Angew. Chem. 1950, 62, 57. 5. Cella, R. J. J. Polym. Sci.: Symp. 1973, 42, 727. 6. Sorta, E. and Melis, A. Polymer 1978, 19, 1153. 7. a) Harrell, L. L. Macromolecules 1969, 2, 607; b) Ng, H. N., Allegrezza, A. E., Seymour, R. W. and Cooper, S. L. Polymer 1973, 14, 255; c) Eisenbach, C. D., Baumgartner, M. and Gunter, G. Adv. Elastomers Rubber Elasticity, [Proc. Symp.] 1985, 51; d) Miller, J. A., Shaow, B. L., Hwang, K. K. S., Wu, K. S., Gibson, P. E. and Cooper, S. L. Macromolecules 1985, 18, 32; e) Shirasaka, H., Inoue, S.-i., Asai, K. and Okamoto, H. Macromolecules 2000, 33, 2776; f) Lai, Y.-C., Quinn, E. T. and Valint, P. L., Jr. J. Pol. Sci. A 1995, 33, 1767; g) Blundell, D. J., Eeckhaut, G., Fuller, W., Mahendrasingam, A. and Martin, C. Polymer 2002, 43, 5197. 8. a) Gaymans, R. J. and Haan, J. L. d. Polymer 1993, 34, 4360; b) Niesten, M. C. E. J., Gaymans, R. J. and Brinke, A. t. Polym. Prepr. 1999, 40, 1012; c) Niesten, M. C. E. J., Gaymans, R. J. and Brinke, A. t. Polym. Prepr. 1999, 40, 1012; d) Niesten, M. C. E. J., Harkema, S., van der Heide, E. and Gaymans, R. J. Polymer 2000, 42, 1131; e) Niesten, M. C. E. J., Tol, R. and Gaymans, R. J. Polymer 2000, 42, 931; f) Niesten, M. C. E. J., Feijen, J. and Gaymans, R. J. Polymer 2000, 41, 8487; g) Niesten, M. C. E. J. and Gaymans, R. J. J. Appl. Polym. Sci. 2001, 81, 1372; i) Niesten, M. C. E. J. and Gaymans, R. J. Polymer 2001, 42, 6199. 9. Peerlings, H. W. I. and Meijer, E. W. Tetrahedron Lett. 1999, 40, 1021. 10. a) Etter, M. C., Urbanczyk-Lipkowska, Z., Zia-Ebrahimi, M. and Panunto, T. W. J. Am. Chem. Soc. 1990, 112, 8415; b) Desiraju, G. R. Organic solid state chemistry; Elsevier: Amsterdam, 1987; Vol. 32; c) Carr, A. J., Melendez, R., Geib, S. J. and Hamilton, A. D. Tetrahedron Lett. 1998, 39, 7447; d) Pathirana, H. M. K. K., Weiss, T. J., Reibenspies, J. H., Zingaro, R. A. and Meyers, E. A. Z. Kristallogr. 1994, 209, 696; e) Perez-Folch, J., Subirana, J. A. and Aymami, J. J. Chem. Cryst. 1997, 27, 367. 11. a) Hanabusa, K., Shimura, K., Hirose, K., Kimura, M. and Shirai, H. Chem. Lett. 1996, 885; b) van Esch, J., De Feyter, S., Kellogg, R. M., De Schryver, F. and Feringa, B. L. Chem. Eur. J. 1997, 3, 1238; c) Loos, M. d., Esch, J. v., Stokroos, I., Kellogg, R. M. and Feringa, B. L. J. Am. Chem. Soc. 1997, 119, 12675; d) van Esch, J., Kellogg, R. M. and Feringa, B. L. Tetrahedron Lett. 1997, 38, 281; e) Van Esch, J., Schoonbeek, F., De Loos, M., Kooijman, H., Spek, A. L., Kellogg, R. M. and Feringa, B. L. Chem. Eur. J. 1999, 5, 937; f) Schoonbeek, F. S., Van Esch, J. H., Wegewijs, B., Rep, D. B. A., De Haas, M. P., Klapwijk, T. M., Kellogg, R. M. and Feringa, B. L. Angew. Chem. Int. Ed. 1999, 38, 1393; g) Schoonbeek, F. S., Van Esch, J. H., Hulst, R., Kellogg, R. M. and Feringa, B. L. Chem. Eur. J. 2000, 6, 2633; h) Brinksma, J., Feringa, B. L., Kellogg, R. M., Vreeker, R. and van Esch, J. Langmuir 2000, 16, 9249; i) De Loos, M., Ligtenbarg, A. G. J., Van Esch, J., Kooijman, H., Spek, A. L., Hage, R., Kellogg, R. M. and Feringa, B. L. Eur. J. Org. Chem. 2000, 3675; j) De Loos, M., Van Esch, J., Kellogg, R. M. and Feringa, B. L. Angew. Chem. Int. Ed. 2001, 40, 613; k)Estroff, L. A. and Hamilton, A. D. Angew. Chem. Int. Ed. 2000, 39, 3447; l) Boileau, S., Bouteiller, L., Laupretre, F. and Lortie, F. New J. Chem. 2000, 24, 845; m) Lortie, F., Boileau, S., Bouteiller, L., Chassenieux, C., Deme, B., Ducouret, G., Jalabert, M., Laupretre, F. and Terech, P. Langmuir 2002, 18, 7218. 12. a) Heikens, D., Meijers, A. and Reth, P. H. v. Polymer 1968, 9, 15; b) Musselman, S. G., Santosusso, T. M., Barnes, J. D. and Sperling, L. H. J. Polym. Sci. 1999, 37, 2586; c) Beck, R. A. and Truss, R. W. Polymer 1999, 40, 307; d) Shirasaka, H., Inoue, S.-i., Asai, K. and Okamoto, H. Macromolecules 2000, 33, 2776; e) Garrett, J. T., Siedlecki, C. A. and Runt, J. Macromolecules 2001, 34, 7066; f) Garrett, J. T., Lin, J. S. and Runt, J. Macromolecules 2002, 35, 161.
63
Chapter 3
13. a) Yilgr, I., Sha'aban, A. K., Steckle, W. P., Jr., Tyagi, D., Wilkes, G. L. and McGrath, J. E. Polymer 1984, 25, 1800; b) Tyagi, D., Yilgr, I., McGrath, J. E. and Wilkes, G. L. Polymer 1984, 25, 1807; c) Yilgr, E., Burgaz, E., Yurtsever, E. and Yilgr, I. Polymer 2000, 41, 849; d) Yilgr, E. and Yilgr, I. Polymer 2001, 42, 7953; e) Yilgr, I., Burgaz, E., Metin, B., Yurtsever, E. and Yilgr, E. Polym. Prepr. 1999, 40, 1126; f) MarcosFernandez, A., Lozano, A. E., Gonzalez, L. and Rodriguez, A. Macromolecules 1997, 30, 3584; g) Mendolia, M. S., Vincenti, P. J., Barr, M. L., Esposito, A., Blum, Y., Schmidt, H.-W., Chen, H. P., Riess, G. and Wu, H.-J. (Colgate-Palmolive Co.) Patent US5919441, 1999. 14. a) Stradella, L. and Argentero, M. Thermochim. Acta 1993, 219, 315; b) Stradella, L. and Argentero, M. Thermochim. Acta 1995, 268, 1. 15. Mark, H. F., Bikales, N. M., Overberger, C. G. and Mengers Encyclopedia of polymer science and engineering; 2 ed.; Wiley-Interscience: New York, 1985. 16. a) Sierra-Vargas, J., Masson, P., Beinert, G., Rempp, P. and Franta, E. Polym. Bull. 1982, 7, 277; b) Asami, R., Takaki, M., Kyuda, K. and Asakura, E. Polym. J. 1983, 15, 139. 17. Hubin, A. J. and Smith, S. (3M Co.) Patent US3436359, 1969. 18. Champion, D. H. and Speranza, G. P. (Texaco Chemical Co.) Patent US5159101, 1992. 19. a) Hest, J. C. M. v. Ph.D. Thesis, Eindhoven University of Technology (Eindhoven), 1996; b) Aert, H. A. M. v. Ph.D. Thesis, Eindhoven University of Technology (Eindhoven), 1997; c) Hirschberg, J. H. K. K. Ph.D. Thesis, Eindhoven University of Technology (Eindhoven), 2001. 20. a) Brown, H. C. Boranes in organic chemistry; Cornell University Press: Ithaca, 1972; b) Brown, H. C., Choi, Y. M. and Narasimhan, S. Synthesis 1981, 605. 21. Vanrenterghem, T., Dubreuil, M. F., Goethals, E. J. and Loontjens, T. J. Polymer International 1999, 48, 343. 22. a) Dubreuil, M. F. and Goethals, E. J. Macromol. Chem. Phys. 1997, 198, 3077; b) Dubreuil, M. F., Farcy, N. G. and Goethals, E. J. Macromol. Rapid Commun. 1999, 20, 383; c) Goethals, E. J., Caeter, P. V., Geeraert, J. M. and Prez, F. E. D. Ang. Makromol. Chem. 1994, 223, 1; d) Oike, H., Yoshioka, Y., Kobayashi, S., Nakashima, M., Tezuka, Y. and Goethals, E. J. Macromol. Rapid Commun. 2000, 21, 1185. 23. Greene, T. W. and Wuts, P. G. M. Protective groups in organic synthesis; 2 ed.; Wiley-Interscience: New York, 1991. 24. Leir, C. M., Hoffman, J. J. and Stark, J. E. (3M Co.) Patent EP0296852, 1998. 25. a) Hrkach, J. S. and Matyjaszewski, K. Macromolecules 1990, 23, 4042; b) Hrkach, J. S. and Matyjaszewski, K. J. Polym. Sci. 1995, 33, 285. 26. Dreyfuss, M. P. and Dreyfuss, P. J. Pol. Sci. A-1 1966, 4, 2179. 27. Flory, P. J. Principles of Polymer Chemistry; Cornell University Press: Ithaca: New York, 1953. 28. a) Coleman, M. M., Sobkowiak, M., Pehlert, G. J., Painter, P. C. and Iqbal, T. Macromol. Chem. Phys. 1997, 198, 117; b) Coleman, M. M., Skrovanek, D. J., Hu, J. and Painter, P. C. Macromolecules 1988, 21, 59; c) Coleman, M. M., Skrovanek, D. J. and Painter, P. C. Makromol. Chem., Macromol. Symp. 1986, 5, 21; d) Coleman, M. M., Lee, K. H., Skrovanek, D. J. and Painter, P. C. Macromolecules 1986, 19, 2149; e) Skrovanek, D. J., Howe, S. E., Painter, P. C. and Coleman, M. M. Macromolecules 1985, 18, 1676.
64
4
Properties and morphology of block copoly(ether urea)s
Abstract The block copoly(ether urea)s with uniform hard blocks consisting of two urea groups that were described in the previous chapter possess elastomeric properties. The crystal structure of a model compound illustrates the formation of long stacks of hydrogen bonded urea groups. Thermal analysis of these polymers demonstrates the reversible melting of the hard blocks, causing the material to flow. The low glass transition temperature ensures excellent low temperature flexibility. The morphology of the material consists of long stacks of associated hard blocks embedded in the soft phase. Elongation of the materials demonstrates their highly elastic behaviour, with a strain at break ranging from 1000 to 2100%. During tensile testing, irreversible deformations and reorganizations of the hard block occur, resulting in a significant amount of tensile set. These well-defined polymers proved to be superior compared to a less-defined analogue possessing a polydisperse hard block.
65
Chapter 4
4.1 Introduction In the previous chapter, the synthesis of block copoly(ether urea)s pTHFy-Ux with uniform hard blocks was described, with x as the number of urea groups in the hard block, and y the number-averaged molecular weight of the pTHF soft block.
O O O m N H N H x n
pTHFy-Ux It was proposed that the hard blocks associate by hydrogen bonding between the urea groups, resulting in the formation of reversible crosslinks (figure 4.1). These physical crosslinks give the material its mechanical strength and elastic behaviour, while the soft, amorphous pTHF phase makes the material flexible.1
Figure 4.1: Schematic representation of a block copolymer with a urea hard block. The number of urea groups in the uniform hard blocks was varied from 1 to 4. An infrared study showed that within the polymers having only one urea group per repeating unit, no strong hydrogen bonds are formed. As a consequence, this polymer has poor mechanical properties, in fact, it is a viscous liquid at ambient temperature. The hard blocks having more than one urea groups associate strongly by hydrogen bonding, inducing elastomeric behaviour for these materials. However, the association between hard blocks comprising three and four urea groups is so strong, that these polymers are hard to process. The polymers having two urea groups in the hard block, show the optimum balance between the strength of hydrogen bonding, and the processibility/solubility. In this chapter, the material properties and morphology of these block copoly(ether urea)s are described. Table 4.1 shows the schematic notation of the different copoly(ether urea)s that were studied. 66
Table 4.1: Copoly(ether urea)s. Copolymer pTHF1100-U pTHF350-U2 pTHF1100-U2 pTHF2000-U2 pTHF2900-U2 pTHF4000-U2b pTHF1100-U3 pTHF1100-U4 Soft blocka pTHF1100 pTHF350 pTHF1100 pTHF2000 pTHF2900 pTHF4000 pTHF1100 pTHF1100
O N H N H N H O N H N H
Hard block
O N H N H
Hard block content (%) 5.0
H N O
H N
33.0 13.5 7.9 5.6 4.1

N H H H N O N
idem idem idem idem

H N O H N O O N H N H _ 2 H N N H O N H H N N H N H
20.6 26.7 13.5
pTHF1100-U2PDc pTHF1100
a) See table 3.1. b) Random copolymer of tetrahydrofuran and ethylene oxide. c) Polydisperse hard block, on average two urea groups.
4.2 Crystal structure of model bisurea. The block copoly(ether urea) with the most interesting properties possess a hard block comprising two urea groups separated by a butylene-spacer; from now on this structural motive is called the bisureido-butylene unit. To gain more insight into the way this bisureaunit associates via hydrogen bonding, model compound 1, comprising the bisureido-butylene unit was synthesized by the reaction of 1,4-diaminobutane with phenylisocyanate (scheme 4.1).
2 NCO + NH2 NH2
O N H N H
H N 1 O
H N
Scheme 4.1: Synthesis of a model compound containing the bisureido-butylene unit.
67
Chapter 4
Crystals suitable for single crystal X-ray analysis were obtained by slow diffusion of water vapour into a solution of 1 in DMSO. In figure 4.2, three molecules in the crystal structure of 1, associated by hydrogen bonding, are shown. More details of this crystal structure are given in table 4.2. The all-trans conformation of the even butylene-spacer, leads to a transoid arrangement of the two urea groups. The urea groups and the least-square plane through the carbon atoms of the butylene-spacer are coplanar. Bifurcated hydrogen bonding between adjacent urea groups gives rise to the formation of an infinite stack of hydrogen bonds. The spacing between two hydrogen bonded urea groups is 4.64 , which is comparable to the distance in crystal structures of substituted urea groups described in literature.2 The angle between the least-square plane through the phenyl-ring and the least-square plane through the urea group is 49.0. In this way, the distance between two superimposed phenyl-rings is 3.24 , and an optimal - stacking between the phenyl-rings in the crystal structure is achieved.3 The distance between to two outer nitrogen atoms N2 and N2 is 10.80 , corresponding to an extremely thin hard block in the block copolymers.4 In general, the schematic picture (figure 4.1) we proposed for very thin, long stacks of hydrogen bonded urea groups corresponds well with this crystal structure. In the direction perpendicular to the plane through the stack of hydrogen bonded urea groups, another stack is situated at a distance of 3.89 . The orientation of the two adjacent stacks is anti-parallel with respect to one another.
Figure 4.2: Single crystal X-ray structure of 1; hydrogen atoms that are not involved in hydrogen bonding are omitted for clarity.
68
Table 4.2: Crystallographic details: hydrogen bond geometry of 1. DH () HA () DA () DHA () N1HO 0.865 (14) 2.075 (14) 2.884 (5) 155.3 (12) N2HO 0.859 (14) 2.186 (14) 2.948 (5) 147.8 (13)
Error between parentheses.
4.3 Thermal properties Optical microscopy Thermoplastic elastomers possess elastic properties at ambient temperatures, and thermoplastic properties at temperatures above the melting point of the thermo-reversible crosslinks. The flow temperatures of the different block copoly(ether urea)s were determined by optical microscopy. The temperatures at which the material lost its dimensional stability, the flow temperature, are depicted in figure 4.3. Since we are interested in the material properties at, and above ambient temperature, the lower limit of the desired flow temperature range is room temperature, and the upper limit is 200C, which is the decomposition temperature of the urea groups.5 This window is indicated by the two dotted horizontal lines. The upper graph shows the flow temperature of the material as a function of the number of methylene units within the aliphatic, linear spacer between the urea groups. The polymers with an even number of methylenes systematically have a higher flow temperature compared to those with an odd number. This is often observed in such a homologous series, including the [n]-nylons and [n]-polyurethanes.6 The polymer with a butylene spacer (n=4) has the highest flow temperature, and this is one of the reasons for us to study this hard segment in more detail. In the left bottom of figure 4.3, the influence of the number of urea groups in the hard block is depicted. As discussed before, the polymer with only one urea group shows viscous flow even below room temperature. Incorporation of one additional urea group makes the material elastomeric and increases the flow temperature to about 140C. The melting of the hard blocks is completely reversible. Increasing the number of urea groups in the hard block to three or four, results in polymers that do not flow below 200C, at which temperature degradation starts to occur. This was evidenced by discolouration and fuming of the sample. Although both materials started to flow when heated slightly above 200C for prolonged times, this process was not reversible, which is also an indication of degradation processes. The polymer with a polydisperse hard block (on average 2 urea groups) shows a flow temperature only slightly lower than the one with exactly two urea groups.
69
Chapter 4
250 200 150

115
pTHF1100-U-CnH2n-U
Tflow (C)
140 126 108 105
100 50 0
n (number of methylene units)
250 200 150 100 50

<20
pTHF1100-Ux
>200 >200
250 200
165
pTHFy-U2
Tflow (C)
Tflow (C)
140
135
150 100 50 0
140 125 102 100
2PD
350
1100
2000
2900
4000
x (number of urea groups)
y (Mn soft block (g/mol) )
Figure 4.3: Flow temperatures of block copoly(ether urea)s. In the right bottom of figure 4.3, the flow temperature as a function of soft block length is shown for the polymers with two urea groups in the hard block. The flow temperature decreases with increasing soft block length. This is a general trend observed for segmented copolymers, and is explained by the solvent effect proposed by Flory.7 Upon dilution of the crystallizable hard segment the size of the crystals will become smaller and hence its melting point decreases. Thus, varying the soft block length enables us to alter the flow temperature of the material.
70
Differential scanning calorimetry Differential scanning calorimetry (DSC) was used to obtain more quantitative data on the thermal transitions of the block copoly(ether urea)s. The second heating and cooling traces of pTHF1100-U2, and pTHF2000-U2 are depicted in figure 4.4, more information is shown in table 4.3.
25 Heat Flow Endo Up (mW) 23 21 19 Tc,h 17 15 -100 Tm,h Tg
Heat Flow Endo Up (mW) 25 23 Tg 21 19 17 15 -100 Tc,h
Tm,s Tm,h
Tc,s
-50
50 T (C)
100
150
200
-50
50 T (C)
100
150
200
Figure 4.4: DSC traces of pTHF1100-U2 (left), and pTHF2000-U2 (right) at a heating/cooling rate of 10C/min. For pTHF1100-U2 no melting- or crystallization peak for the pTHF-block is observed, it is completely amorphous. Its molecular weight is too low to crystallize, as is often observed in segmented block copolymers possessing pTHF as one of the segments.1,4,8 The amorphous soft phase shows a glass transition temperature of 68C, which is close to the glass transition temperature of pure pTHF9, indicating the absence of phase mixing between hard and soft phases. This complete phase separation is advantageous for the low temperature properties of the material. A broad melting endotherm with a maximum at 131C of the hard block is observed. This temperature is close to the flow temperature of 140C, which was determined by optical microscopy. Upon cooling, the hard block crystallized at 102C. Increasing the soft block length results in semicrystallinity of the pTHF phase, as is demonstrated by the DSC of pTHF2000-U2, in which a melting peak at 1C and a crystallization peak at 27C is observed. Table 4.3 shows that the crystallinity and melting temperature of the pTHF phase increase with increasing soft block length (series pTHF350-U2 to pTHF2900-U2). Random incorporation of ethylene oxide units in the pTHF chain successfully reduces the crystallinity and melting point of the soft phase, resulting in a completely amorphous soft phase at room temperature. The melting points of the hard blocks correlate well with the flow temperatures, and the same trend is observed of decreasing 71
Chapter 4
melting points with increasing soft block lengths. No melting peaks for the polymers with three of four urea groups in the hard blocks were observed below 200C. The polymer with the polydisperse hard block shows two melting endotherms, in contrast two the polymers with a uniform hard block. Table 4.3: Thermal transitions of block copoly(ether urea)s as determined by DSC; Subscript s: soft block, h: hard block, -: not observed. Copolymer pTHF1100-U pTHF350-U2 pTHF1100-U2 pTHF2000-U2 pTHF2900-U2 pTHF4000-U2 b pTHF1100-U3 pTHF1100-U4 pTHF1100-U2PD Tga (C) 79 68 74 73 70 67 69 Tm,s (C) 9 1 19 1 4 Hm,s (J/g) 27.2 20.3 38.3 34.9 12.8 Tc,s (C) 15.8 27 10 -37 23 Tm,h (C) 158 131 118 78 1) 114 2) 142 Hm,h (J/g) 20.0 14.2 3.9 3.6 3.0 0.73 Tc,h (C) 145 102 101 48 81
a) Determined at a heating rate of 40C/min. b) Random copolymer of tetrahydrofuran and ethylene oxide.
Infrared spectroscopy The extent of hydrogen bonding upon heating the material was studied by temperature-dependent infrared spectroscopy. A polymer film was cast from chloroform solution on a potassium bromide pellet, and spectra were recorded in transmission. Figure 4.5 shows the carbonyl region infrared spectra at several temperatures of polymer pTHF2000-U2. According to the previously used techniques, the melting temperature of the urea based hard block is between 118 and 125C. However, at 130C the infrared spectrum still strongly resembled the spectrum at room temperature with a peak corresponding to strong hydrogen bonding. Further heating of the sample resulted in an increase of contributions due to weakly hydrogen bonded and non-hydrogen bonded urea groups.10 Obviously, strong hydrogen bonds between the urea groups are still present in the melt, although the material lost its dimensional stability. Heating the sample to 200C for prolonged times resulted in an infrared spectrum in which a new peak at 1730 cm-1 dominates. This peak is indicative of degradation of urea groups11, and this process is irreversible. 72
200C
140C 130C
%T
120C 20C
1800
1700
-1 (cm )
1600
1500
Figure 4.5: Temperature-dependent infrared spectra of pTHF2000-U2. Dynamic mechanical thermal analysis Dynamic mechanical thermal analysis (DMTA) was used to study the temperature dependency of the material properties. The storage modulus and tan of pTHF1100-U2 are shown in figure 4.6. The maximum in tan at 50C indicates a glass transition, at which the polymer goes from a rigid, glassy material to a rubbery material. From 5 to 110C, there is an extended, temperature-independent rubbery plateau, (E=135 MPa). The constant value of this storage modulus is an indication that phase separation is complete and no phase transitions occur within this temperature range. This is an advantage compared to other thermoplastic elastomers that often show a decrease of the rubbery plateau with increasing temperature, due to incomplete phase separation or partially melting of the hard blocks. Slightly above 120C, the hard blocks start to melt, and the material begins to soften. It finally collapses at 148C. These temperatures correspond well with those found with DSC and optical microscopy. In table 4.4 some characteristics of other copoly(ether urea)s are depicted. Not all polymers are included in this table, since for some polymers it was impracticable to perform DMTA measurements. The storage moduli of these materials at 25C decrease with increasing soft block length, i.e. the materials become softer with decreasing hard block content.1,4 The difference in modulus between polymers pTHF1100-U2 and pTHF2000-U2 is remarkably high. A possible explanation for this behaviour is the existence of a co-continuous 73
Chapter 4
hard phase for pTHF1100-U2 compared to a disperse hard phase for pTHF2000-U2.12 Similar to the results found with optical microscopy and DSC, a decrease of flow temperature with increasing soft block length was observed by DMTA.
10 9.5 0.4 9 8.5 8 7.5 7 6.5 6 -100 0 200 0.2 0.3 tan 0.1 -50 0 50 T (C) 100 150 0.5
Figure 4.6: DMTA curves of pTHF1100-U2, measured at 1 Hz and a heating rate of 1C/min. Table 4.4: Characteristics of block copoly(ether urea)s as determined by DMTA; Subscript s: soft block, -: not observed, n.d.: not determined. Copolymer pTHF1100-U2 pTHF2000-U2 pTHF2900-U2 pTHF4000-U2 pTHF1100-U3 pTHF1100-U2PD Tg (C) 50 n.d. n.d. 62 n.d. n.d. Tm,s (C) n.d. n.d. n.d. n.d. E(25C) (MPa) 135 17 n.d. 11 66 n.d. Tflow (C) 148 112 n.d. 105 >200 n.d.
4.4 Morphology The morphology of the block copoly(ether urea)s, as proposed by us, is schematically depicted in figure 4.1. Long stacks of hydrogen bonded urea groups, i.e. the hard blocks, are 74
log E' (Pa)
embedded in the amorphous pTHF phase. In this way, reversible crosslinks are formed, which give the material its elastomeric properties. To examine the validity of this model, the morphology of pTHF2000-U2 was studied by AFM and X-ray diffraction techniques. Atomic force microscopy Thin films of approximately 30 nm thickness were cast on a silicon wafer from chloroform/methanol solution. These substrates were subsequently heated to 150C and slowly cooled to room temperature, to allow slow crystallization of the hard blocks. The structures of these films were studied by AFM in the tapping mode. Figure 4.7 shows the AFM images of polymer pTHF2000-U2, both in height and phase contrast. The resolution of the latter is better. The hard phase appears lighter than the soft phase in this image. Long fibers of the hard block embedded in the soft phase are visible. The length of these fibers ranges to over 500 nm, and the apparent width is estimated at 10 nm. However, smaller details cannot be discerned, since the resolution is limited by the sharpness of the AFM-tip.
Figure 4.7: AFM images of pTHF2000-U2; height (left) and phase contrast (right). Wide-angle X-ray diffraction Wide-angle X-ray diffraction (WAXD) and small-angle X-ray scattering (SAXS), have been used extensively to investigate the morphology of thermoplastic elastomeric block copolymers.1,12,13 Cella proposed the nowadays commonly accepted lamellar morphology for segmented copolymers, in which the crystalline phase consists of hard segments, embedded in a mixed amorphous phase, which consists of the soft segments and uncrystallized hard segments.14 Bonart has confirmed this domain structure by SAXS.15
75
Chapter 4
For the present block copoly(ether urea)s, WAXD and SAXS measurements were carried out on non-oriented solution-cast samples of pTHF1100-U2. Figure 4.8 shows the WAXD profile of this polymer at room temperature. A broad peak is visible, originating from the amorphous pTHF phase. No reflections are observed that reveal the presence of a crystalline pTHF phase.16 On top of the broad WAXD peak, a reflection is superimposed at s = 0.226 -1, (d-spacing of 4.42 ). This corresponds well with the spacing between the hydrogen bonded urea groups (4.6 ), that was found in the crystal structure of model compound 1 (figure 4.2). Due to the low hard block content of this polymer (13.5%), this reflection is rather small.
1.2 1 0.8 I(s) 0.6 0.4 0.2 0 0.1 0.15 0.2 0.25 0.3
-1
0.35
0.4
)sin ( ) s = (2/) )
Figure 4.8: WAXD curve of non-oriented solution-cast sample of pTHF1100-U2. Small-angle X-ray scattering Figure 4.9 (left) shows the SAXS curve of the same polymer. The analysis of the small-angle X-ray scattering intensities, I(s), was based on the linear correlation function:
K1 (r ) = I (s ) s 2 cos(2rs ) ds
0
(eq. 4.1)
that was obtained via cosine transformation of the data.17 It is important to note that this approach is based on a two-phase, lamellar morphology in which the width of the domains is much larger than their height. The analysis assumes periodic stacking of crystalline and amorphous domains with a long period, L, and lamellar thickness of the crystalline domain, d.
76
0.100
1.0
0.075
0.050
(r) = K(r)/Q
0.5
lamellar thickness d ~ 15
I(s)*s
0.0
long period L ~ 60
0.025
0.000 0.00
-0.5
0.01 0.02 0.03 -1 s = (2/)*sin ( ) 0.04 0.05
50 r ()
100
150
Figure 4.9: Onedimensional scattering intensities I1(s) obtained from SAXS experiments
(left); and the normalized linear correlation function, (r), calculated from these data (right). The linear correlation function (figure 4.9 right) reveals the long period, L=60 , and the lamellar thickness, d=15 . Despite the limitations of the applied model, these structural parameters correspond very well to the typical dimensions of the polymer architecture. According to the crystal structure of model compound 1, the distance between the carbon atoms on the outside of both urea groups is only slightly less than 15 (figure 4.10). d
O N H O N H O N H O N H O N H O N H O N H N H N H N H N H N H N H N H H N O H N O H N O H N O H N O H N O H N O H N H N H N H N H N H N H N O N H O N H O N H O N H O N H O N H O N H N H N H N H N H N H N H N H H N O H N O H N O H N O H N O H N O H N O H N H N H N H N H N H N H N
L Figure 4.10: Schematic picture of the morphology of the block copolymer; not all soft blocks
are plotted for clarity. The lamellar model assumes crystalline domains of infinite size in two dimensions. According to AFM, which shows very long hydrogen bonded stacks, one dimension can indeed be considered as infinite. However, in the second dimension perpendicular to the former one, the size of the crystalline domain can not be infinite, otherwise plate-like structures would be observed in AFM. Another indication for the limited size is the crystallinity of the sample as estimated from to the SAXS measurements. A long period of 60 and a lamellar thickness of 15 , implies a crystallinity of 25 volume%. However, based on the ratio between hard and soft blocks, the crystallinity is expected to be much lower, to be 77
Chapter 4
exact 13.5 weight%. This suggests a pseudo two-phase morphology in which crystalline domains of rather limited size are dispersed in the amorphous matrix.18 Such a morphology would give rise to a diffuse scattering intensity at very low angles. However, our measurements do not allow for such an analysis, due to the limited amount of data points in this region. Real-time SAXS/WAXD/DSC measurements The melting and order-disorder transition of block copolymers has been studied by temperature-dependent SAXS and/or WAXD experiments.19 For many thermoplastic elastomers, it is a question whether the crystallization of the hard blocks starts from a homogenous melt, or from a pre-ordered (phase-separated) melt. The latter possibility represents a rather attractive aspect, since crystallization of the hard blocks would use the existing structure as a template. To examine the development of the microstructure of the material upon heating and cooling, time-resolved temperature-dependent combined SAXS/WAXD measurements were performed on beamline BM26B at the ESFR. Figure 4.11 shows the different SAXS and WAXD profiles at increasing temperature, and clearly reveals the melting of the structure. Above the temperature at which the material starts to flow (140C), no structural features were observed in the SAXS pattern, indicating a homogenous melt state in which the hard blocks are dissolved in the soft phase. In the WAXD pattern, the reflection superimposed on top of the broad amorphous pTHF peak, decreases upon heating, and eventually disappears completely, demonstrating the melting of the stacks of hydrogen bonded urea groups. For SAXS, the total scattering power, Q, which is defined by: Q = I ( s ) * s 2 ds
0
(eq. 4.2)
is plotted as a function of temperature in figure 4.12 (left). Both a heating and a cooling run are depicted. The heating curve first shows a gradual decrease of the scattering power. At approximately 105C, a more rapid decrease was observed, indicating the onset of melting of the ordered structure. For the WAXD, the patterns were deconvoluted into contributions from amorphous and ordered material and the peak areas of the diffraction corresponding to the stacked urea groups were extracted, and plotted as a function of the temperature (figure 4.12 right). The comparison of the evolution of the SAXS and the WAXD intensities reveals a simultaneous decrease of both signals, which suggests the melting and dissolution of the hard domains. The temperature at which this transition occurs is slightly lower than the temperature at which the modulus starts to decrease, as observed by DMTA (120C). On the other hand, it corresponds 78
well with the onset of melting that was observed by DSC (ca. 100C). Apparently, partially melting of the structure does not immediately result in a decrease of the modulus.
1.0x10
-4
5.0x10
-5
T(C) 25 37.5 50 62.5 75 87.5 100 112.5 125 137.5 150
I(s) 0.0 0.000
0.005
0.010 0.015 0.020 -1 s = (2/)*sin ( )
0.025
0.030
0.3
T(C) 25 37.5 50 62.5 75 87.5 100 112.5 125 137.5 150
0.2 I(s)
0.1
0.0 0.10
0.15
0.20
0.25
0.30
-1
0.35
0.40
s = (2/)*sin ( )
Figure 4.11: SAXS (top) and WAXD (bottom) profiles during heating of pTHF1100-U2.
79
Chapter 4
0.20
1.6x10 Scattering power Q (a.u.)
-6
Diffraction peak area (a.u.)
1.2x10
-6
0.15
heating
8.0x10
-7
cooling
0.10
4.0x10
-7
0.05
0.0 -50
50 T (C)
100
150
0.00 -50
50 T (C)
100
150
Figure 4.12: Scattering power, Q, as a function of temperature, during heating and subsequent
cooling at a rate of 5C/min. In figure 4.12 (left), some hysteresis is observed upon heating and cooling the sample. An undercooling of 15C is observed, which is smaller than that observed by DSC (29C). This discrepancy is partly explained by the different cooling rates employed in both techniques: 5C/min for the SAXS, and 10C/min for DSC, respectively. The undercooling is rather small compared to other polymers, like polyethylene, poly(butylene terephtalate), or polyamides,20 indicating a high rate of crystallization.
4.5 Mechanical properties
Tensile testing Tensile and elastic properties of the block copoly(ether urea)s with different soft block lengths were studied. Films of these polymers were cast from chloroform/methanol solution, and this procedure yielded completely transparent, flexible films. Tensile bars were punched from these non-oriented films, and examined by tensile testing. Figure 4.13 and table 4.5 show the stressstrain behaviour of these materials. The strain, , is expressed as the ratio between the elongation (L) and the initial length (L0), in terms of percentage.
80
40
30
pTHF1100-U2
pTHF2000-U2
(MPa)
20
pTHF1100-U2PD pTHF4000-U2
10
0 0 500 1000 (%) 1500 2000 2500
Figure 4.13: Stressstrain curves for block copoly(ether urea)s. Table 4.5: Tensile properties of block copoly(ether urea)s.
Copolymer
Youngs Yield stress Strength Strain Toughness modulus at break E (MPa) y (MPa) br (MPa) br (%) (kJ/kg) 95.8 88.3 25.6 11.3 9.6 6.7 4.7 2.4 27.7 15.3 27.8 11.0 1060 805 1175 2140 178 80 147 122
pTHF1100-U2 pTHF1100-U2PD pTHF2000-U2 pTHF4000-U2
The stressstrain curves are typical for thermoplastic elastomeric materials. At small deformations, stress increases linearly with strain (Hookean behaviour). The Youngs modulus of these polymers increases with decreasing soft block length (increasing hard block content). According to Wegner21, the logarithm of the modulus of a segmented copolymer follows a linear relationship with the volume fraction of crystallinity. In our case, when log E is plotted versus the hard block content (see table 4.1), the data are confirming this empirical relationship, at least for these four points (figure 4.14).
81
Chapter 4
2.5
log E (MPa)
1.5
0.5
0 0 5 10 Hard block content (%) 15
Figure 4.14: Relation between Youngs modulus and hard block content.
As the stress is increased above the yield stress, plastic deformation occurs. At this stage, the stressstrain curve deviates considerably from Hookean behaviour, since stress is redistributed by deformation (fragmentation) and reorganization of hard segments. Finally, the polymer cannot bear the load anymore, and the material breaks. The strain at break of these polymers is surprisingly high, ranging from 1000 to 2100%. Harrell22 and Miller23 already observed high strains at break for segmented copolyurethanes having a narrow hard segment length distribution. This is caused by the facile deformation and reorganization of thin crystalline lamellae, so the stress is more evenly distributed over the soft segments.4 The steep upswing of the stressstrain curve (strain hardening) of pTHF2000-U2 is caused by strain induced crystallization of the pTHF soft segment. This is evidenced by a whitening of the material. Upon releasing the stress, the soft phase remains semicrystalline, as was proved by DSC. Although pTHF1100-U2 shows a strain hardening effect as well, no strain induced crystallization is observed, probably since the molecular weight of the soft block is too low to crystallize. Remember, pTHF4000 is a random copolymer of THF and ethylene oxide, and this successfully prevents strain induced crystallization. The toughness is the energy that is absorbed until the material fails, and is reflected by the area under the stressstrain curve. Due to the high strengths and strains at break for these materials, the materials are much tougher than many other thermoplastic elastomers.1 The toughness of spider dragline silk (approx. 123 kJ/kg)24 is even exceeded, however, this is not a fair comparison, since spider dragline silk is highly oriented and in this way predeformed. 82
The block copoly(ether urea) with the polydisperse hard block, pTHF1100-U2PD possesses inferior properties compared to the analogous one with the uniform hard block, pTHF1100-U2. The Youngs modulus, strength, strain at break, and toughness are considerably lower for the former polymer, demonstrating the importance of the well-defined molecular structure for the material properties. Cyclic tensile testing An important feature of elastomers is the ability to relax to their original dimensions after being deformed. However, complete elastic recovery is rarely observed for thermoplastic elastomers; they usually show some tensile set after being elongated. This is caused by irreversible changes in the morphology of the material during deformation. Once the material has been plastically deformed, the second initial modulus is lower compared to the original initial modulus. In general, segmented copolymers with a co-continuous hard phase show more tensile set than those with a disperse hard phase.12 Furthermore, strain induced crystallization increases the tensile set. The extent of tensile set of the copoly(ether urea)s was determined by performing cyclic tensile tests. The specimen were prestrained to a certain amount of strain, subsequently unloaded and allowed to relax, and again elongated to the same strain (figure 4.15).
30 25 20 15 10
3 2 and 4
(MPa)
5 0 0 200 400 600 (%) 800 1000 1200
Figure 4.15: Cyclic tensile test of pTHF1100-U2, testing speed is 100%/min.
83
Chapter 4
The tensile set is defined as:
TS =
L1 L0 *100% L0
(eq. 4.3)
with L0 as the initial length of the specimen, and L1 as the length after straining and relaxing. A significant amount of hysteresis is observed for these polymers. In figure 4.16 (left), the tensile set is plotted as a function of the prestrain for pTHF1100-U2 and pTHF4000-U2. Per polymer two curves are depicted: one immediately after relaxation, and one after 24 hours of relaxation. The polymer with the longer soft block, pTHF4000-U2, shows less tensile set than that with the shorter soft block, pTHF1100-U2: a tensile set after 100% prestrain (TS100%) of 6% and 34%, respectively, is observed immediately after deformation. Relaxation of the specimen for 24 hours reduces the tensile sets to 4.5% and 25%, respectively. A possible explanation for this behaviour is the existence of a co-continuous hard phase for pTHF1100-U2 compared to a disperse hard phase for pTHF4000-U2.12 The tensile sets of these materials, especially pTHF1100-U2, are rather high compared to natural rubber (TS100%<2%), but comparable or lower than those of other thermoplastic elastomeric materials, e.g. block copoly(ether ester)s (Arnitel, TS100%=3540%)12, block copoly(ether aramid)s (TS300%=150 170%)4, or spandex (TS100%=25%)25. However, the latter comparison is not completely fair, since this value concerns a solution-spun spandex fiber, which is highly oriented, and in this way predeformed. Plastic deformation is also quantified by the hysteresis energy, which is the percentage of energy dissipation within a loadingunloading cycle (difference between areas under the loading and unloading curves).26,28c When the hysteresis energy is 0%, all the work (energy) that is put into the sample to deform it, is recovered after relaxation (ideal rubber).
800
pTHF1100-U2, t=0 pTHF1100-U2, t=24h pTHF4000-U2, t=0 pTHF4000-U2, t=24h
100 90 80 Hysteresis energy (%) 70 60 50 40 30 20 10 pTHF4000-U2 pTHF1100-U2
600 Tensile set (%)
400
200
0 0 400 800 1200 Prestrain (%) 1600
0 0 400 800 1200 Prestrain (%) 1600
Figure 4.16: Left: tensile set of two polymers immediately after prestraining, and after 24
hours; right: hysteresis energies of these polymers as a function of prestrain.

84
On the other hand, a hysteresis energy of 100% means that all the work is used to induce irreversible changes in the material, and no work is recovered upon relaxation, the energy is lost as heat. For both polymers, the relative hysteresis energies are plotted in figure 4.16 (right). At prestrains above the yield point, almost all energy used to elongate the material is dissipated, resulting in a high amount of tensile set. This is caused by the facile disruption of the hard segments. Once the material has been deformed, the tensile behaviour is almost completely reversible (compare traces 2 and 4 in figure 4.15). This situation, more or less, resembles the behaviour in highly oriented fibers, like solution-spun spandex fibers.28c
4.6 Orientation during tensile testing
In this section, we investigated the changes in morphology and structure during deformation of the block copoly(ether urea)s. Upon elongation (straining) of the native material, orientation and irreversible changes in the morphology of the material occur, resulting in plastic deformation. The orientation in polymers can be measured by several techniques.27 We decided to use infrared linear dichroism spectroscopy and X-ray scattering techniques to study the structural changes within the material during tensile deformation. Infrared linear dichroism The orientation of functional groups within a polymer is often studied by infrared linear dichroism spectroscopy.28 In this technique, the absorption of a functional group is measured with polarized infrared light at an angle parallel (A) and perpendicular (A) to the stretching direction. The dichroic ratio for such a group is defined as: R = A / A (eq. 4.4) For uniaxial orientation, the degree of orientation is expressed by the Hermans orientation function: ( R 1)( R 0 2) 3cos 2 1 (eq. 4.5) f = = ( R 0 1)( R 2) 2 in which is the angle between the main chain axis and the deformation axis, and R0 is the dichroic ratio for perfect uniaxial orientation, and is given by: (eq. 4.6) in which is the angle between the transition dipole moment of the absorbing group and the main chain axis. The values of are either described in literature28c, or determined from the crystal structure in figure 4.2, and are given in table 4.6. A distinction is made between peaks 85
R 0 = 2 cot 2
Chapter 4
arising from the hard, and from the soft segments. The first two absorptions are corresponding to the urea groups in the hard segment, and the last absorption to the polyether soft segment.
Table 4.6: Peak assignments of selected infrared bands of block copoly(ether urea)s.
Peak Position (cm-1) 3325 1615 1110
Assignment NH stretching C=O stretching COC stretching
Block hard hard soft
() 90 78 0
For randomly oriented groups, f=0; for groups at which the main chain is aligned parallel to the deformation axis, f=1; and for perpendicular alignment, f=0.5. Figure 4.17 shows the orientation functions of the different functional groups of polymer pTHF1100-U2 with respect to the deformation axis as a function of elongation , up to an elongation of 700%. Subsequently, the specimen was released and allowed to relax for several minutes (right side of figure 4.17).
1.00 NH C=O
0.75 Orientation function f (-)
0.50
0.25
COC
0.00 0 -0.25 100 200 300 400 500 600 700 relaxed
-0.50
(%)
Figure 4.17: Orientation functions of functional groups in pTHF1100-U2.
As expected, the orientation functions of the three groups are zero before stretching the sample, confirming the non-oriented state of the sample (figure 4.18a). The graph shows that at low strain levels, below the yield point, the chain axis in the hard segment is oriented 86
transverse to the deformation axis (f<0). Since the hydrogen bonded stacks of urea groups are perpendicular to the chain axis, this means that these stacks are oriented parallel to the deformation axis (figure 4.18b), which is caused by the high aspect ratio of these hard domains. At these low strains, hardly any orientation of the polyether soft segment is observed. Above the yield point of the polymer, a different deformation mechanism sets in. Now, the chain axis in the hard segments becomes oriented parallel to the deformation axis (f>0), meaning that the stacks of hydrogen bonded urea groups (the hard domains) become oriented perpendicular to the deformation axis. To allow this, fragmentation of the hard domains occurs, permitting the reorientation of the hydrogen bonding stacks (figure 4.18c).26,28 Upon elongation, the polyether soft block is also oriented, as is reflected by the orientation function of ether COC stretching vibration. This indicates orientation of the polymer main chain (and thus the soft segment) in the direction of the deformation. Releasing the stress on the sample results in relaxation of the soft block to a random coil conformation. No residual orientation of the polyether segment is observed (f=0). In contrast, there is still a high amount of residual orientation of the hard segments, which is responsible for the tensile set of the material.
H N
H N
H N
H N
H N
H N
H N O
N H
H N
N H
O N H O N H O N H O N H N H N H N H N H
N H
H N
N H
H N
H N O H N
O H N
O N H O N H O N H O N H N H H N O N H H N O H N N H H N O N H H N O
N H N H N H N O O N H O H N
H N
H N
H N H N O
N H N
H N H N O
N H N
H N
H N H N O
N H N O H
O
H
H N H N O
N H N O H
H N H N O
N H N O H
H N H
H N O
O N
N H N O H
N H N O H
H N
O N H N H O N H N H O
N O
H N
H N
N H
H N
H N
H N
N H O
H N
H N
N H
N H
H N
N H
O O N H
N H
H N
N H
N H
N H
H N
H N
H N
H N
H N
H N
H N
N H
N H
H N
N H
N H
H N
H N
H N
H N
H N
N H
N H
H N
N H
N H
N H
N H
H N
H N
N H
N H
N H
N H
N H
H N
N H O N H H N O H N O
N H O N H H N O
N H O N H H N O
N H O N H H N
N H
H N
N H
H N
N H
H N
N H
N H
N H
N H
N H
N H O
N H
N H
N H
N H
N H
H N N H O
H N
H N O H N
H N
H N
H N
N H
N H
N H
N H
N H
N H
N H
N H
N H
H N N H O
O N H
O H N N H H N O
N H
Figure 4.18: Schematic representation of molecular orientation during tensile testing; a) non-
oriented, b) below yield point, c) beyond yield point; not all soft blocks are plotted for clarity. This orientation behaviour upon elongation is comparable to what is described in literature for other segmented thermoplastic elastomers.
N H
H N O H N H N H N O H N H N O
N H H N O
H N
H N
H N
H N
H N
H N
H N O O H N
H N
O H N H N
H N
H N
N H N H O N H N H N H H N O H N O O O N H N H N H
H N O
O H N H N O
H N H N O H N H N
N H
H N
87
Chapter 4
2-Dimensional wide-angle X-ray diffraction X-ray scattering techniques, both WAXD and SAXS, are very suitable to study orientation of polymeric materials.13f,27,28a,f,29. Especially, a combination of these techniques with microscopic techniques (e.g. AFM) is very powerful.30 The orientation of polymer pTHF1100-U2 during tensile testing was studied by timeresolved 2D-WAXD measurements at ID11 beamline at the ESRF, using a miniaturized tensile tester provided by DSM Research. A typical result, obtained at a strain of 500% is shown in figure 4.19 (right). Only two reflections (each others mirror images) are visible on the equator. Due to the low crystallinity of the material, and the overlap of the reflection corresponding to the hard block with the amorphous pTHF halo (see figure 4.8), no conclusions can be drawn concerning the orientation of the distinct segments in this polymer upon elongation. Nevertheless, since these reflections are observed on the equator, perpendicular to the deformation axis, the overall orientation within the elongated specimen should be perpendicular to the deformation axis, as well (figure 4.18c). Strain induced crystallization is not detected, since no reflection corresponding to a crystalline pTHF phase are observed.16 Relaxation of the applied stress results in partial recovery of the amorphous halo (figure 4.19 right). Some remaining orientation is observed, as the reflections on the equator are still visible.
deformation axis
Figure 4.19: 2D-WAXD patterns of pTHF1100-U2 at 500% strain (left), and after relaxation
for 15 min (right).
88
2-Dimensional small-angle X-ray diffraction The orientation and reorganization of the different segments in block copoly(ether urea) pTHF1100-U2 was studied using 2D-SAXS measurements on beamline BM26B at the ESFR. Due to the absence of a tensile tester suitable for these measurements, the samples were prestrained to a specific deformation, allowed to relax for two days, and subsequently measured. Figure 4.20 shows three SAXS patterns (only the lower half) at different amounts of prestrain. The circular spot in the middle of the patterns is a defect due to damaging of the detector. The pattern at 25% prestrain reveals a non-oriented morphology due to full reversibility of the microstructure. At 150% prestrain a four-spot pattern, and at 800% prestrain a two-spot pattern is observed.
25%
150%
800%
Figure 4.20: 2D-SAXS patterns recorded from pTHF1100-U2 after prestraining to specified
deformation, deformation axis is vertical. The four-spot pattern develops upon deformations higher than 100%, and is observed up to deformations of 500%. It indicates that there are correlations among the hard domains not only parallel, but also perpendicular to the deformation axis. We attribute this four-spot pattern to the tilted formation of the hard blocks along the deformation axis (figure 4.18b). This type of orientation of hard blocks in segmented copolymers has been reported before by Bonart15 and Hsiao31, and is caused by the high aspect ratio of these domains. The two-spot pattern is superimposed on the four-spot pattern after deformations higher than 500%, and it dominates the scattering pattern above 800%. It indicates the presence of rather small, fragmented, crystalline domains that are ordered in stacks with correlations among the stacks only in the direction parallel to the deformation axis, and not in the lateral direction (figure 4.18c). Further analysis of this two-spot pattern, by plotting the intensities recorded perpendicular to the deformation axis in a Guinier plot32, revealed an 89
Chapter 4
average dimension of the hard domains after deformation of approximately 5 nm. This is significantly smaller than the length of the hard domains as found by AFM (figure 4.7), which reached over 500 nm. This decrease is due to the fragmentation of the hard domains at high deformation values. It would be highly interesting to study the fragmentation of the hard domains upon elongation in real-time by X-ray scattering techniques. However, attempts to do so were unsuccessful so far. This was caused by the low crystallinity of the material, and the limited thickness of the tensile bars that were used for this purpose.
4.7 Conclusions
The block copoly(ether urea)s with uniform hard blocks possess elastomeric properties when the hard blocks contain two urea groups or more. The polymers with exactly two urea groups in the hard blocks, pTHFy-U2, show the most interesting properties, since they are highly elastic, and start to flow below the decomposition temperature polymer. The crystal structure of a model compound, 1, comprising the same bisureido-butylene unit, illustrated the intermolecular hydrogen bonding geometry between the urea groups, giving rise to long stacks forming the hard domain. Optical microscopy, DSC, DMTA, and variable temperature FT-IR, WAXD and SAXS proved that above the melting temperature of the hard domains, the hydrogen bonds between the urea groups are broken, and the material starts to flow. This process is completely reversible. The flow temperature of these materials ranged from 100 to 165C, and decreased with increasing soft block length. No phase mixing between hard and soft phases was observed at room temperature, and this complete phase separation results in an extended, temperature-independent rubbery plateau and in a low glass transition temperature, and hence good low temperature flexibility. The morphology of the copolymers was studied by AFM and X-ray scattering techniques. AFM clearly illustrated the presence of very long stacks of hydrogen bonded urea groups embedded in the soft pTHF phase. This morphology resembles that of a nanocomposite. Based on a two-phase lamellar morphology, a long period of 60 and a lamellar thickness of 15 were determined by SAXS. These dimensions correspond well to the molecular picture. Tensile testing showed that these polymers behave as typical thermoplastic elastomers, with a Youngs modulus that depends on the hard block content. Strain at break for these polymers ranges from 1000 to 2100%. The molecular orientation during tensile testing was studied by infrared linear dichroism spectroscopy and X-ray scattering techniques. 90
Two deformation processes were observed. Below the yield point, the hard domains tilt in a direction parallel to the deformation axis due to their high aspect ratio. In contrast, above the yield point, fragmentation and reorientation of the hard domains occurs, resulting in perpendicular orientation. Upon releasing the stretched material, the soft segment relaxes to the non-oriented state, but the hard domains remain oriented. This leads to a considerable amount of tensile set. These well-defined polymers proved to be superior compared to a less-defined analogue possessing a polydisperse hard block. Compared to other thermoplastic elastomers, the material properties of these novel block copoly(ether urea)s are equal or better. The strain at break and toughness of these materials are higher than for many commercially available TPEs, e.g. block copolyesters (trade name Hytrel or Arnitel), block copolyamides (Pebax, Vestamid), or block copolyurethanes (Desmopan, Elastane). Another advantage is that these block copoly(ether urea)s are completely transparent, in contrast to many of the other TPEs. We propose that the unique properties of these polymers are the result of the very thin, uniform hard blocks and the strong hydrogen bonding between the urea groups within these hard blocks. To make a direct comparison with spandex fibers, it is necessary to spin fibers from these materials. The low flow temperature of these polymers might allow melt spinning to produce the highly oriented, elastic fibers. This procedure would be advantageous with respect to the more expensive solution spinning that is currently used for the fabrication of spandex fibers.28c Until so far, we have demonstrated the synthesis and detailed characterization of these well-defined block copolymers possessing uniform hard blocks. We are now at a stage that we can extend this concept in two ways. First, we would like to utilize the uniformity of the hard blocks for the modification of the material properties by a modular approach (chapter 5). And secondly, we would like to extend the concept to well-defined block copolymers in which also the soft block is monodisperse (chapter 6).
4.8 Experimental section
General methods and instrumentation. General methods concerning purification of solvents, spectroscopic techniques and mass spectrometric techniques can be found in the experimental section of previous chapters. Optical microscopy: Optical properties and flow temperatures were determined using a Jeneval polarization microscope equipped with a Linkam THMS 600 heating device with crossed polarizers. Temperature-dependent infrared spectroscopy: Infrared spectra were measured on a Perkin Elmer 1600 FTIR. Films of the polymer were cast in a potassium bromide pellet, and heated in a Linkam THMS 600 heating device.
91
Chapter 4
Differential scanning calorimetry: DSC was performed on a Perkin DSC 7, melting and crystallization temperatures were determined at a heating/cooling rate of 10C min1, glass transition temperatures at a heating rate of 40C/min. Dynamic mechanical thermal analysis: DMTA spectra were recorded on a Polymer Laboratories Mk III Dynamic Mechanical Analyzer. A small rectangular bar of the polymer, size approximately 12x10x0.15 mm were subjected to a sinusoidal deformation at constant frequency, using the tensile method. Measurements were carried out at a frequency of 1 Hz and a heating rate of 1C/min. The amplitude of the sinusoidal deformation was 10m, and the static force was 0.2 N in order to ensure good contact between sample and probe. Atomic force microscopy: AFM experiments were performed using a NanoScope III A instrument (Digital Instruments) operating in the tapping mode, utilizing NanoSensor tapping tips. The amplitude of oscillation at free vibration, A0, was set to 4.0 V. The operating setpoint ratio (A/A0) was set to relatively low values (A/A0 ~ 0.7) to obtain best contrast. For visualization of the phase separation in thin copolymer films (typically 1 or 2 m scan sizes), the phase image gave best contrast. Small-angle X-ray scattering: The small angle X-ray scattering (SAXS) experiments were performed at the SAXS station of the Dutch-Belgium beamline BM26b of the European Synchrotron radiation facility (ESRF) at Grenoble, using an X-ray wavelength of 0.082nm and a sample-to-detector distance of about 4m. Twodimensional scattering data were recorded with a Gabriel-type multiwire detector having 512*512 pixels, each with a size of 300*300m. The detector was placed in an off-center position in order to cover a broad range of scattering vectors from 0.002 < s < 0.04, where s = (2/)*sin (with , the wavelength and 2, the scattering angle). Individual patterns were normalized with respect to the primary beam intensity and corrected for sample absorption with the aid of ionization chambers placed before and after the sample position. Two-dimensional scattering pattern from oriented samples were recorded from stacks of the pre-strained material in order to obtain an acceptable signal-to-noise ratio. From these patterns a transmission weighted background was subtracted and the resulting profiles for samples with deformations > 150% were analyzed further since in that case a fiber symmetry can be assumed -see P2() parameter from WAXD experiments. In this case, integrations through selected areas in reciprocal space parallel and perpendicular to the orientation axis were performed. Temperature-dependent profiles during heating/cooling cycles were recorded in a temperature range from 25160C (rate 5C/min), using a remote controlled LINKAM DSC stage. In this case, linear scattering profiles were extracted by performing a sector integration over the two-dimensional patterns, followed by a subtraction of the transmission-weighted background scattering. The final calibration of the patterns with respect to the scattering vector was based on the diffraction maxima observed for a wet rat-tail collagen (d ~ 65 nm). Initial data processing including normalization to primary beam intensity and detector response as well as background subtraction and sector integrations to obtain linear profiles were performed with DUBBLE software, based on the BSL program. Further evaluation including sections through selected areas of the 2D patterns were performed with the FIT2D software. Processing of the linear profiles including calculation and evaluation of the linear correlation functions was performed with the OTOKO software. Tensile testing: Stressstrain measurements were performed on a Zwick Z010 Universal Tensile Tester at an elongation rate of 100%/min. Tensile bars were punched from a solution-cast film of the polymers. Dimensions of the tensile bars: length=22 mm, width=5.0 mm, and thickness=0.30 mm (approx.). Infrared linear dichroism spectroscopy: Infrared (IR) spectra were recorded by means of a Biorad UMA 500 microscope, coupled to an FTS6000 FT-IR spectrometer. Films of the polymer with a thickness of 10 m were cast in a Teflon mould, and tensile bars (22*5.0 mm2) were punched from this. The specimen was elongated and clamped. Infrared spectra were recorded in transmission at 0 and 90 polarization angle with respect to the deformation axis.
92
1,4-butanediyl-bis(N-phenylurea), 1 1,4-Diaminobutane (2.00 g, 22.7 mmol) was dissolved in chloroform (30 ml), and phenylisocyanate (5.94 g, 50.0 mmol) in chloroform (20 ml) was added dropwise. The suspension was stirred for 1 h, filtered, washed with chloroform., and dried. The product was obtained as a white, microcrystalline powder (7.10 g, 96%). Tm=243C (Lit: 240C). 1H-NMR (DMSO-d6): 8.39 (s, 2H, PhNH), 7.38 (dd, 4H, C2H-phenyl, Jortho=8.2 Hz, Jmeta=0.9 Hz), 7.21 (m, 4H, C3H-phenyl, J=7.7 Hz), 6.87 (t, 2H, C4H-phenyl, J=6.8 Hz), 6.14 (t, 2H, CH2NH, J=5.7 Hz), 3.11 (m, 4H, CH2NH), 1.45 (m, 4H, CH2CH2CH2CH2). 13C-NMR (DMSO-d6): 155.2 (C=O), 140.6 (C1phenyl), 128.6, (C3-phenyl), 120.9 (C4-phenyl), 117.6 (C2-phenyl), 38.8 (CH2N), 27.3 (CH2CH2CH2CH2). FTIR (ATR): 3320 (NH stretching), 2940, 2865, 1626 (C=O stretching), 1595, 1556, 1225, 732 cm-1. Anal. Calcd. (%) for C18H22N4O2: C 66.24, H 6.79, N 17.17. Found (%): C 66.24, H 6.22, N 17.07. X-ray Crystal Structure Analysis. Single crystals suitable for X-ray analysis were obtained by slow diffusion of water-vapor into a 50 g/l solution of the compound in DMSO. C18H22N4O2, Mr = 326.40, colourless, plate-shaped crystal (0.1 0.3 0.3 mm), cut from a larger crystal, monoclinic, space group C2/c (no. 15) with a = 35.975(4), b = 4.646(8), c = 9.8781(10) , = 99.846(10), V = 1627(3) 3, Z = 4, Dx = 1.333 g cm3, F(000) = 696, (MoK) = 0.090 mm1. 15525 Reflections measured, R = 0.0264, 1849 independent reflections, Rint = 0.0572, 1.0 < < 27.45, T = 150 K, MoK radiation, graphite monochromator, = 0.71073 . Data were collected on an Enraf-Nonius KappaCCD area detector on rotating anode. The structure was solved by direct methods (SHELXS86), and refined on F2 using SHELXL-97-2. Positional parameters of the hydrogen atoms were refined; initial values were obtained from a difference Fourier map. All non-hydrogen atoms were refined with anisotropic thermal parameters; hydrogen atoms were refined with a fixed isotropic thermal parameter related to the value of the equivalent isotropic displacement parameter of their carrier atoms by a factor of 1.2. Final refinement of 142 parameters resulted in a wR2-value of 0.1067, w = 1 / [(Fo2) + (0.0642 P)2 + 0.70 P], where P = (max(Fo2,0) + 2 Fc2) / 3, R1 = 0.0373 for 1598 I > 2 (I), S = 1.030, 0.23 < < 0.22 e 3.

1. Legge, N. R., Holden, G. and Schroeder, H. E. Thermoplastic elastomers: A comprehensive review; Carl Hansser Verslag: New York, 1987. 2. a) Born, L. and Hespe, H. Colloid Polym. Sci. 1985, 263, 335; b) van Esch, J., De Feyter, S., Kellogg, R. M., De Schryver, F. and Feringa, B. L. Chem. Eur. J. 1997, 3, 1238; c) Van Esch, J., Schoonbeek, F., De Loos, M., Kooijman, H., Spek, A. L., Kellogg, R. M. and Feringa, B. L. Chem. Eur. J. 1999, 5, 937; d) De Loos, M., Ligtenbarg, A. G. J., Van Esch, J., Kooijman, H., Spek, A. L., Hage, R., Kellogg, R. M. and Feringa, B. L. Eur. J. Org. Chem. 2000, 3675; e) Gesquiere, A., Abdel-Mottaleb, M. M. S., De Feyter, S., De Schryver, F. C., Schoonbeek, F., van Esch, J., Kellogg, R. M., Feringa, B. L., Calderone, A., Lazzaroni, R. and Bredas, J. L. Langmuir 2000, 16, 10385; f) Etter, M. C., Urbanczyk-Lipkowska, Z., Zia-Ebrahimi, M. and Panunto, T. W. J. Am. Chem. Soc. 1990, 112, 8415; g) Carr, A. J., Melendez, R., Geib, S. J. and Hamilton, A. D. Tetrahedron Lett. 1998, 39, 7447; h) Pathirana, H. M. K. K., Weiss, T. J., Reibenspies, J. H., Zingaro, R. A. and Meyers, E. A. Z. Kristallogr. 1994, 209, 696; i) Perez-Folch, J., Subirana, J. A. and Aymami, J. J. Chem. Cryst. 1997, 27, 367. 3. a) Hunter, C. A. and Sanders, J. K. M. J. Am. Chem. Soc. 1990, 112, 5525; b) Jorgensen, W. L. and Severance, D. L. J. Am. Chem. Soc. 1990, 112, 4768; c) Chipot, C., Jaffe, R., Maigret, B., Pearlman, D. A. and Kollman, P. A. J. Am. Chem. Soc. 1996, 118, 11217. 4. Niesten, M. Ph.D. Thesis, University of Twente (Enschede), 2000. 5. a) Stradella, L. and Argentero, M. Thermochim. Acta 1993, 219, 315; b) Stradella, L. and Argentero, M. Thermochim. Acta 1995, 268, 1. 6. a) Baeyer, A. Ber. Chem. Ges. 1877, 10, 1286; b) Boese, R., Weiss, H.-C. and Blaser, D. Angew. Chem. 1999, 38, 988; c) Thalladi, V. R., Boese, R. and Weiss, H.-C. Angew. Chem. 2000, 39, 918; d) Thalladi, V. R., Boese, R. and Weiss, H.-C. J. Am. Chem. Soc. 2000, 122, 1186; e) Thalladi, V. R., Nsse, M. and Boese, R. J. Am. Chem. Soc. 2000, 122, 9227; f) Aharoni, S. M. n-Nylons: Their synthesis, structure, and properties; J. Wiley &
93
Chapter 4
Sons: Chichester, 1997, pp 60; g) Versteegen, R. M., Sijbesma, R. P. and Meijer, E. W. Angew. Chem. Int. Ed. 1999, 38, 2917; h) section 2.4 of this thesis. 7. Flory, P. J. Trans. Faraday Soc. 1955, 51, 848. 8. Olabisi, O. Polyester-based thermoplastic elastomers; Marcel Dekker: New York, 1997, pp 1053. 9. Mark, H. F., Bikales, N. M., Overberger, C. G. and Mengers Encyclopedia of polymer science and engineering; 2 ed.; Wiley-Interscience: New York, 1985. 10. Coleman, M. M., Sobkowiak, M., Pehlert, G. J., Painter, P. C. and Iqbal, T. Macromol. Chem. Phys. 1997, 198, 117. 11. Saunders, J. H. and Frisch, K. C. Polyurethanes : chemistry and technology, part 1 Chemistry; Interscience: New York, 1962; Vol. 16. 12. Schmalz, H., van Guldener, V., Gabrieelse, W., Lange, R. and Abetz, V. Macromolecules 2002, 35, 5491. 13. a) Garrett, J. T., Lin, J. S. and Runt, J. Macromolecules 2002, 35, 161; b) Garrett, J. T., Runt, J. and Lin, J. S. Macromolecules 2000, 33, 6353; c) Savelyev, Y. V., Akhranovich, E. R., Grekov, A. P., Privalko, E. G., Korskanov, V. V., Shtompel, V. I., Privalko, V. P., Pissis, P. and Kanapitsas, A. Polymer 1998, 39, 3425; d) Musselman, S. G., Santosusso, T. M., Barnes, J. D. and Sperling, L. H. J. Polym. Sci. 1999, 37, 2586; e) Veenstra, H., Hoogvliet, R. M., Norder, B. and De Boer, A. P. J. Polym. Sci. B 1998, 36, 1795; f) Chu, B. and Hsiao, B. S. Chem. Rev. 2001, 101, 1727. 14. Cella, R. J. J. Polym. Sci.: Symp. 1973, 42, 727. 15. a) Bonart, R. J. Macromol. Sci. B 1968, 2, 115; b) Bonart, R. and Mller, E. H. J. Macromol. Sci., Phys 1974, B10, 177; c) Bonart, R. and Mller, E. H. J. Macromol. Sci., Phys 1974, B10, 345. 16. Cesari, M., Perego, G. and Mazzei, A. Makromol. Chem. 1965, 83, 196. 17. Strobl, G. R. and Schneider, M. J. Pol. Sci. B 1980, 18, 1343. 18. Goderis, B., Reynaers, H. and Koch, M. H. J. Macromolecules 2002, 35, 5840. 19. a) Phillips, R. A. and Cooper, S. L. Macromolecules 1995, 28, 5734; b) Velankar, 2000 #445]; c) Heck, B., Arends, P., Ganter, M., Kressler, J. and Stuehn, B. Macromolecules 1997, 30, 4559. 20. a) Koutsky, J. A., Walton, A. G. and Baer, E. J. Appl. Phys. 1967, 38, 1832; b) van Bennekom, A. C. M. and Gaymans, R. J. Polymer 1997, 38, 657. 21. Wegner, G., In Legge, N. R., Holden, G. and Schroeder, H. E. Thermoplastic elastomers: A comprehensive review; Carl Hansser Verslag: New York, 1987, Section 12/5. 22. Harrell, L. L. Macromolecules 1969, 2, 607. 23. Miller, J. A., Shaow, B. L., Hwang, K. K. S., Wu, K. S., Gibson, P. E. and Cooper, S. L. Macromolecules 1985, 18, 32. 24. Kubik, S. Angew. Chem. Int. Ed. 2002, 41, 2721. 25. Ullmann Ullmann's Encyclopedia of industrial chemistry; 6 ed.; VCH: Weinheim, 2001. 26. Wang, C. B. and Cooper, S. L. Macromolecules 1983, 16, 775. 27. Ward, I. M. Structure and properties of oriented polymers; 2 ed.; Chapman & Hall: London, 1997. 28. a) Van der Heide, E., Van Asselen, O. L. J., Ingenbleek, G. W. H. and Putman, C. A. J. Macromol. Symp. 1999, 147, 127; b) Lee, H. S. and Hsu, S. L. J. Pol. Sci. B 1994, 32, 2085; c) Lee, H. S., Ko, J. H., Song, K. S. and Choi, K. H. J. Polym. Sci. B 1997, 35, 1821; d) Kischel, M., Kisters, D., Strohe, G. and Veeman, W. S. Eur. Pol. Journal 1998, 34, 1571; e) Graff, D. K., Wang, H., Palmer, R. A. and Schoonover, J. R. Macromolecules 1999, 32, 7147; f) Lee, H. S., Ko, J. H., Song, K. S. and Choi, K. H. J. Polym. Sci. B 1997, 35, 1821. 29. Blundell, D. J., Eeckhaut, G., Fuller, W., Mahendrasingam, A. and Martin, C. Polymer 2002, 43, 5197. 30. Sauer, B. B., McLean, R. S., Brill, D. J. and Londono, D. J. J. Polym. Sci. B 2002, 40, 1727. 31. Yeh, F., Hsiao, B. S. and Sauer, B. Polym. Mater. Sci. Eng. 1998, 79, 332. 32. a) Schultz, J. M., Lin, J. S. and Hendricks, R. W. J. Appl. Crystallogr. 1978, 11, 551; b) Tan, S., Zhang, D. and Zhou, E. Polymer International 1997, 42, 90.
94
5
A modular approach to polymer modification and functionalization
Abstract The uniformity of the bisureido-butylene hard block of the block copoly(ether urea)s enabled us to supramolecularly anchor to these polymers molecules containing the same unit. Increasing the spacer-length between the urea groups disturbs this recognition, and prevents successful intercalation. This concept of supramolecular intercalation was used to increase the hard block content of the block copolymers, leading to higher flow temperatures, and stiffer materials. Furthermore, the dye disperse orange 3 was functionalized with the complementary unit, and after mixing with the polymer, it was strongly anchored to the hard block. This dye could not be removed by washing. In contrast, the dye lacking the complementary unit was easily washed out, since its association is much weaker. By elongating the polymer containing the functionalized dye, this dye was highly oriented, as observed with UV-Vis dichroism spectroscopy. This again confirms the tight supramolecular interaction between the dye and the hard block of the polymer.
95
Chapter 5
5.1 Introduction In the past century, with the explosive growth of the world of plastics, polymer scientists have expanded the number of polymers for specific applications enormously. In order to meet the demand of polymeric materials with special properties and functions, numerous monomers have been synthesized, and a variety of polymerization and processing techniques have been developed. This way of making tailor-made polymers is costly and time-consuming, since for every novel polymer, new monomers and/or polymerization techniques must be developed. It would be highly beneficial to obtain these specialty polymers via a modular approach.1 In such a strategy, a limited set of building blocks is used to obtain materials with a variety of properties and/or functionalities, by cleverly putting together the components. A next step, to broaden the scope of the modular approach, would be to specifically use supramolecular interactions. Employing non-covalent interactions between the building blocks, permits a method in which the functional components are simply added to the polymeric host, to which they subsequently adhere via self-assembly. This allows multiple functional materials to be obtained from a single, polymeric backbone, in a divergent manner. Recently, Rotello et al. discussed the reversible functionalization of a polystyrene backbone via hydrogen bonding between dipropamidopyridone functionalized side chains and a guest flavin.2 They studied the interaction between the guest and polymeric host both in solution and in bulk, which demonstrated the highly efficient recognition process. They foresee that this non-covalent plug and play approach opens the way towards a modular methodology of polymer modification. In our systems, we like to use the nature of the hard block to reversibly functionalize polymeric materials. Due to its uniformity and specific length between the ureas, we anticipated on employing this structural element as a recognition site for the reversible binding of guest molecules. Hydrogen bonding between the bisureido-butylene unit in the hard block of the polymers, and a guest possessing a complementary bisureido-butylene should result in a specific, supramolecular interaction, causing the guest to be bound tightly within the hard block of the polymer (scheme 5.1). In this way, modification of the polymer properties, or functionalization of the material can be effectuated in a modular approach, by simply mixing the complementary guests with the polymers, providing a new methodology for material development.
96
O N H O N H N H N H
H N O H N O
H N N H
O N H O N H O N H N H N H
H N O H N O H N O
H N
H N
H N
Functional group
+
O N H N H H N O H N
H N
Functional group
Scheme 5.1: Schematic representation of reversible functionalization of polyureas.
5.2 Increase of hard block content Introduction In a first approach, to examine the validity of the plug and play principle, the idea of a reinforcing filler was employed. In polymer technology, fillers are widely used to alter the properties of polymers. Many types of filler are known, either organic or inorganic in origin.3 Our idea was to prepare two compounds that both contain two urea groups, however, they have a different spacer in-between. Compound 1 contains a butylene-spacer, with a binding motif that is complementary to the hard block of polymer pTHF-U2, in contrast to 2 which has a pentylene-spacer. These compounds can then be mixed with pTHF-U2, so that 1 intercalates in the hard block of the polymer, whereas 2 cannot, since it does not fit properly. By doing so, 1 should increase the hard block content and, therefore, act as a reinforcing filler at the molecular level.
O C6H13 N H N H 1 H N O H N C6H13 O C6H13 N H N H 2 N H O N C6H13 H
Results and discussion Compounds 1 and 2 were prepared by reaction of hexyl isocyanate and 1,4diaminobutane and 1,5-diaminopentane, respectively. Although both compounds are poorly soluble in pure chloroform, they are reasonably soluble in mixtures of chloroform and methanol. Compound 1 was dissolved together with polymer pTHF1100-U2 in a mixture of methanol (25 v%) in chloroform (75 v%). The solution was poured into a Petri-dish, and the 97
Chapter 5
solvent was allowed to evaporate slowly until complete dryness. This was also done for the mixture of bisurea 2 and pTHF1100-U2. The first combination gave a transparent film, while on the contrary, the film containing 2 was turbid and clearly showed macrophase separation. This is the first indication that 1 does fit in the hard block of the block copolymer, while 2 does not. The amounts of 1 and 2 that were mixed with the polymer were varied from 0 to about 20 w%. Figure 5.1 shows the increase of flow temperature of the mixtures as a function of the amount of filler. For the film containing 1, an increase of the flow temperature was observed of 35C upon addition of 20 w% of the filler. The melt of this material was completely homogenous. In contrast, the flow temperature of the film containing 2 increased to a much lower extent (only 10C). This melt was not homogenous, but showed the presence of isolated crystallites of 2 floating in the viscous polymer phase. Apparently, these crystallites do not reinforce the material.
180 1 170
Tflow(C)
160
150
140
130 0 5 10 15 20 25 filler (w%)
Figure 5.1: Flow temperature versus concentration of fillers 1 and 2. In figure 5.2, the flow temperature is plotted as a function of the hard block content for the pure pTHFy-U2 (table 4.1 and figure 4.3 right bottom) and for polymer pTHF1100-U2 reinforced with filler 1. An extra large increase of the flow temperature upon addition of the filler is observed, compared to the expectations based on the pure block copoly(ether urea)s.
98
180 170 160 150 Tflow(C) 140 130 120 110 100 0 5 10 15 20 25 30 35 hard block content (w%)
Figure 5.2: Flow temperature as a function of hard block content for pure pTHFy-U2 (), and for pTHF1100-U2 filled with 1 (). The successful intercalation of filler 1 is also reflected in the stiffness of the material: addition of 5 w% of 1 increased the modulus by 50% according to DMTA.
5.3 Incorporation of dye molecules Introduction The intercalation of complementary compounds into the block copoly(ether urea)s was extended to molecules bearing functionalities. A dye was selected as the functionality, since its nature can be easily monitored by optical spectroscopy. Many examples are known of polymers containing dyes, especially azobenzene moieties.4 They were studied for e.g. their mechanochromic behaviour, photoinduced orientation, topographical nanostructure patterning, or non-linear optical (NLO) properties. All these examples involve the covalent attachment of the chromophore to the polymer, either to the backbone, or to a side chain. Smith et al. studied the orientation of dyes and -conjugated polymers that were blended with polyolefins.5 Subsequently drawing these films resulted in highly oriented chromophores, with dichroic ratios exceeding 20. In our approach, the dye is incorporated in a non-covalent manner, by molecular recognition between the polymer and the dye.
99
Chapter 5
Synthesis Disperse orange 3, a well known dye, was derivatized to compounds possessing one (4), or two (5) urea groups using a route given in scheme 5.2. The branched, racemic 2-ethylhexyl chain was introduced to increase the solubility of the modified dyes.
O2N N N NH2 COCl2 O2N N 3 O 3 + H2N O2N N N 4 N H N H N NCO
Disperse Orange 3
O 3 + H2N NBoc H O2N O TFA O2N N N N H N H NH2 + OCN N N N H N H NHBoc
O O2N N N N H N H 5
H N O
H N
Scheme 5.2: Modified dye molecules containing urea groups. Washing of films containing the modified dyes Dye 5 possesses the complementary bisureido-butylene unit, and is expected to fit in the hard block of the copoly(ether urea)s, in contrast to 4 which has only one urea group. To test this hypothesis, films were prepared of polymer pTHF1100-U2 containing both compounds in a concentration of 3 w%, by dissolving both components in a chloroform-methanol mixture and casting this solution into a Petri-dish. This gave rise to two red, transparent, elastic films. Using optical microscopy, macrophase separation was observed in neither of the two films. Square pieces of 1 cm2 were cut from these films, and these were individually stirred in an aqueous solution of 0.1 M sodium dodecylsulfate (SDS) at 60C for 90 minutes. This washing procedure had a remarkably different effect on the pieces of polymer film. The film 100
containing the compound with only one urea group, 4, discoloured rapidly. After 90 minutes, it had become pale, while the washing water had an intense red colour, Indicating that the dye is easily solubilized because it is loosely bound in the polymer. In contrast, the piece of film containing dye 5 having the complementary bisureidobutylene unit kept its red colour. Even after prolonged washing, the washing water remained colourless, although the isolated dye, itself, is readily solubilized in the aqueous solution. This proves that the latter dye is strongly anchored to the hard block of the polymer, whereas the former dye is not and thus is easily washed out.
Figure 5.3: Films and aqueous phases after washing the films containing the modified dyes 4 (left) and 5 (right). Elongation of films containing the modified dyes In section 4.6 it was shown that upon elongation of the block copolymer, the hard domains are highly oriented in the direction perpendicular to the deformation axis (figure 4.18c). If dye 5 is indeed strongly anchored to the hard block of the polymer, then it should also become oriented when the film is stretched. This would result in an anisotropy of the chromophore, that is observable by UV-Vis dichroism spectroscopy. Films of polymer pTHF4000-U2 containing dyes 4 and 5 were elongated and the dichroic ratio (D.R.) was determined. A typical UV-Vis spectrum of a film containing dye 5 that is elongated by a factor of 5 (=500) is shown in figure 5.4. The maximum absorption is at 405 nm. This graph clearly shows the difference in absorption depending on the polarization angle of the light, and a dichroic ratio of 10.3 is estimated at 405 nm.
101
Chapter 5 0.5 =0 =90 0.4
0.3 Abs 0.2 0.1 0 350
400
450
(nm)
500
550
600
Figure 5.4: UV-Vis spectra of stretched film with dye 5 measured at two angles of polarized light (=0: parallel and =90: perpendicular to the deformation axis). In figure 5.5, the dichroic ratio is plotted as a function of the elongation of the film. Due to the experimental setup it was difficult to elongate the material more than 5 times. As becomes obvious in figure 5.5, the dichroic ratio of the film containing the complementary dye 5 increased rapidly, indicating that the chromophore of the dye is highly oriented upon stretching the film. In contrast, the dichroic ratio of the dye having only one urea group, 4, increased barely, with a maximum dichroic ratio of only 1.8, demonstrating that this dye is weakly oriented in the polymer. This is in agreement with the assumption that the dye 5 containing the bisureidobutylene-unit is anchored to the hard block of the thermoplastic elastomer and this hard block is highly oriented upon stretching the material (see section 4.6). On the other hand, dye 4 is not incorporated in the hard block, but it is dissolved in the amorphous phase, which is much less oriented. A schematic presentation of the orientation of the dye upon elongation is shown in figure 5.6. The material is elongated along the z-axis, and the hard blocks are oriented perpendicular to this axis. This causes the dye (and its dipole moment) to orientate along the z-axis (=0) as well, resulting in a higher absorption of light at this polarization angle.
102
A modular approach to polymer modification and functionalization 12 5
10
8 D.R.
0 0 100 200 300 (%) 400 500 600
Figure 5.5: Dichroic ratio as a function of the elongation of the dyed films.
NO2
H N
H N
H N
H N H N O O N H N H H N O
N H N H N H N O N H H N O N H N H N H O N H N H O
Figure 5.6: Schematic representation of dye 5 in the hard block of an elongated film.
5.4 Conclusions The block copoly(ether urea)s, which have been described in the previous chapters, possess uniform hard blocks comprising urea groups. This enables us to incorporate guest molecules containing the complementary bisureido-butylene unit. Via hydrogen bonding 103
N H
N H
Chapter 5
between the urea groups of the guest, and the hard block of the polymer, the guests are strongly associated. Increasing the length of the spacer between the ureas disturbs the supramolecular recognition, in this way preventing the incorporation. Mixing the polymers with filler 1 resulted in an increase of the hard block content, and hence, an increase of the flow temperature and modulus of the material. Therefore, by mixingin complementary compounds, the material properties of the block copolymers can be altered. A modified dye molecule, bearing the bisureido-butylene unit, 5, was mixed with the polymer, and compared to a dye containing only one urea group, 4. The former was much stronger anchored to the polymer. While the latter dye was easily washed out off a piece of polymer film, the former one remained in the polymer. Stretching of the film containing the complementary dye resulted in a strong orientation of the chromophore, as was observed by UV-Vis dichroism spectroscopy. Dye 4 was hardly oriented upon elongation of the film. This difference is explained by the fact that dye 5 is incorporated in the hard block, which is highly oriented upon stretching. In contrast, dye 4 residing somewhere in the amorphous phase, is much less oriented. We demonstrated a novel approach of non-covalent functionalization of polymers by simply mixing-in complementary guest molecules. This concept opens the way towards a modular methodology of polymer modification. We have only demonstrated this with two examples, but the number of functionalities that can be incorporated is almost unlimited. For example, the incorporation of catalytic sites opens the opportunity to catalytically active materials of which the activity can be changed by mechanical deformation of the material.

General methods and instrumentation. General methods concerning purification of solvents and spectroscopic techniques can be found in the experimental section of previous chapters. UV-Vis spectra were recorded on a Perkin Elmer Lambda 900 spectrometer, equipped with a polarizer accessory. The dichroic ratio was measured at max, by placing the polarizer either parallel or perpendicular to the deformation axis, Materials. The availability of most chemicals was discussed in the experimental section of previous chapters. Disperse Orange 3 was purchased from Aldrich; phosgene (20% in toluene) and 2-ethylhexyl amine were purchased from Fluka. Preparation of filled films. PTHF-U2 (ca. 2 g) and the desired amount of filler-compound were dissolved in chloroform (15 ml) and methanol (5 ml). This solution was cast in a silylated Petri-dish (diameter 9 cm), and the solvent was allowed to evaporate slowly by placing a beaker over the dish. After 20 h, the film was dried in vacuo at 50C for 5 h, and it was peeled off the Petri-dish.
104
1,4-butanediyl-bis(N-hexylurea), 1 1,4-Diaminobutane (0.83 g; 9.36 mmol) was dissolved in chloroform (40 ml), and to this solution was added hexyl isocyanate (2.50 g; 19.66 mmol) in chloroform (20 ml). The solution turned turbid within minutes, and the reaction mixture was stirred for 30 min, and subsequently filtered. The product was washed with chloroform, recrystallized from a 1:1 methanol/chloroform mixture (150 ml) and dried in vacuo. It was obtained as a white powder (2.63 g, 82%). Tm=220C. 1H-NMR (DMSO-d6, 120C): 5.52 (br, 4H, NH), 2.98 (t, 8H, CH2NH), 1.41 (m, 8H, NCH2CH2), 1.32 (m, 12H, (CH2)3CH3), 0.88 (t, 6H, CH3). FT-IR (ATR): 3321 (NH stretching), 2931, 2858, 1614 (C=O stretching), 1567, 1476, 1221 cm-1. 1,5-pentanediyl-bis(N-hexylurea), 2 1,5-Diaminopentane (0.96 g; 9.36 mmol) was dissolved in chloroform (40 ml), and to this solution was added hexyl isocyanate (2.50 g; 19.66 mmol) in chloroform (20 ml). The turbid reaction mixture was stirred for 30 min, and subsequently filtered. The product was washed with chloroform, recrystallized from a 1:1 methanol/chloroform mixture (100 ml) and dried in vacuo. It was obtained as a white powder (2.66 g, 80%). Tm=189C. 1H-NMR (DMSO-d6): 5.72 (br, 4H, NH), 2.95 (t, 8H, CH2NH), 1.35-1.21 (br.m, 22H, CH2CH2C), 0.86 (t, 6H, CH3). FT-IR (ATR): 3326 (NH stretching), 2932, 2859, 1611 (C=O stretching), 1564, 1475, 1253, 1217 cm-1. 4-Isocyanato-4-nitroazobenzene, 3 Disperse Orange 3 (4-(4-Nitro-phenylazo)-aniline) (0.50 g, 2.07 mmol) was dissolved in THF (40 ml), and phosgene (2.2 ml 20% in toluene, 4.1 mmol) was added. The reaction mixture was heated to reflux temperature and stirred for 1 h, while argon was bubbled through the solution. It was evaporated to dryness, and the product was obtained as a red solid (0.62 g, 112%). FT-IR (ATR): 2257 (NCO), 1734 (NHCOCl). cm-1. 3-(2-Ethyl-hexyl)-1-[4-(4-nitro-phenylazo)-phenyl]-urea, 4 4-Isocyanato-4-nitroazobenzene, 3, (0.31 g, 1.00 mmol) was dissolved in THF (15 ml), and 2-ethylhexylamine (0.20 g, 1.5 mmol) in THF (5 ml) was added. The reaction mixture was stirred at room temperature for 30 min, after which it was evaporated to dryness. The product was redissolved in chloroform (20 ml), and extracted with hydrochloric acid solution (10 ml 0.1 M in water), and saturated sodium bicarbonate solution (10 ml). The organic layer was dried with sodium sulfate, filtered and purified by column chromatography using 1% methanol in chloroform as the eluent (Rf = 0.4 ). The product was obtained as an orange solid (0.30 g, 75%). 1H-NMR (DMSO-d6): 9.02 (s, 1H, Ph-NH), 8.41 (d, 2H, C2H, J=9.2 Hz), 8.01 (d, 2H, C3H, J=8.8 Hz), 7.91 (d, 2H, C2H, J=8.8 Hz), 7.65 (d, 2H, C3H, J=8.8 Hz), 6.35 (t, 1H, CH2NH), 3.08 (q, 2H, CH2NH, J=5.9 Hz), 1.40 (m, 1H, CH), 1.28 (m, 8H, C-CH2-C), 0.89 (t, 6H, CH3, J=6.2 Hz). FT-IR (ATR): 3336 (NH stretching), 2960, 2928, 1669 (C=O stretching), 1595, 1543, 1515, 1340, 1226, 1140, 1105, 859, 843, 685 cm-1. 3-(2-Ethyl-hexyl)-1-(3-[4-(4-nitro-phenylazo)-phenyl]-ureido-1,4-butyl)-urea, 5 4-Isocyanato-4-nitroazobenzene, 3, (0.55 g, 2.07 mmol) was dissolved in THF (30 ml), and 4-(Tertbutoxycarbonylamino)-1-butylamine (0.58 g, 3.11 mmol) in THF (4 ml) was added. The reaction mixture was stirred at room temperature for 30 min, after which it was partially concentrated and precipitated in pentane (100 ml). The product was filtered off, and purified by column chromatography using 1% methanol in chloroform as the eluent (Rf=0.3). It was redissolved in dichloromethane (3 ml), and trifluoroacetic acid (2 ml) was added to deprotect the amine group. The reaction mixture was stirred at room temperature overnight, and subsequently evaporated to dryness. Di-tert-butyl tricarbonate (0.40 1.54 mmol) was dissolved in chloroform (10 ml), and 2-ethylhexyl amine (0.19 g, 1.47 mmol) in chloroform (2 ml) was injected into the former solution. The reaction mixture was stirred for 30 min. The previously deprotected amine was dissolved in pyridine (50 ml), and added to the former solution. The reaction mixture was stirred for 30 min at room temperature, and then evaporated to dryness. The product was purified by column chromatography, first using pure chloroform as the eluent, than chloroform-methanol mixtures with up to 10% methanol (Rf = 0.2). The product was obtained as an orange solid (0.28 g, 27%). 1HNMR (10% methanol-d4 in CDCl3): 8.37 (d, 2H, C2H, J=8.1 Hz), 7.99 (d, 2H, C3H, J=8.4 Hz), 7.93 (d, 2H,
105
Chapter 5
C2H, J=8.4 Hz), 7.59 (d, 2H, C3H, J=9.2 Hz), 3.26 (t, 2H, PhNHCONHCH2), 3.15 (t, 2H, PhNHCONHCH2CH2CH2CH2), 3.07 (t, 2H, NHCONHCH2 CH), 1.54 (m, 4H, NHCH2CH2CH2CH2NH), 1.4-1.2 (m, 9H, CH + CH2), 0.88 (t, 6H, CH3). FT-IR (ATR): 3322 (NH stretching), 2924, 2859, 1633 + 1623 (C=O stretching), 1584, 1552, 1523, 1343, 1226, 1140, 1106, 865, 754 cm-1. UV-Vis (THF): max=405 nm.

1. Michl, J. Modular Chemistry; Kluwer: Dordrecht, 1997. 2. Ilhan, M. F., Gray, M. and Rotello, V. M. Macromolecules 2001, 34, 2597. 3. a) Struik, L. C. E., Bree, H. W. and Schwarzl, F. R. Proc. Int. Rubber Conf., 5th 1968, 205; b) HiljanenVainio, M., Heino, M. and Seppala, J. V. Polymer 1998, 39, 865. 4. a) Mller, M. and Zentel, R. Macromolecules 1996, 29, 8024; b) Xin, Z., Sanda, F. and Endo, T. J. Polym. Sci. 2001, 39, 2620; c) Hasegawa, M., Ikawa, T., Tsuchimori, M., Watanabe, O. and Kawata, Y. Macromolecules 2001, 34, 7471; d) Tawa, K., Kamada, K., Kiyohara, K., Ohta, K., Yasumatsu, D., Sekkat, Z. and Kawata, S. Macromolecules 2001, 34, 8232; e) Buffeteau, T., Labarthet, L. F., Pezolet, M. and Sourisseau, C. Macromolecules 2001, 34, 7514; f) Shin, Y.-D., Ahn, J.-H. and Lee, J.-S. Polymer 2001, 42, 7979. 5. a) Weder, C., Sarwa, C., Bastiaansen, C. and Smith, P. Adv. Mater. 1997, 9, 1035; b) Montali, A., Bastiaansen, C., Smith, P. and Weder, C. Nature 1998, 392, 261; c) Palmans, A. R. A., Eglin, M., Montali, A., Weder, C. and Smith, P. Chem. Mater. 2000, 12, 472.
106
6
Block copolymers containing a monodisperse soft block
Abstract Monodisperse pTHF-oligomers, 1-(x), were successfully synthesized via a two step synthetic cycle. A 13-mer (molecular weight of 954 g/mol) was prepared on a 20 gram scale, and a detailed analysis showed that it had a polydispersity of approximately 1.004. This oligomer was used as a monodisperse soft block in a block copoly(ether urea), pTHFMD-U2, possessing a hard block with two urea groups. Dissolving this polymer in chloroform resulted in the formation of a gel. This behaviour is substantially different from the analogous polymer with a polydisperse soft block, pTHF1100-U2,which is highly soluble in chloroform. After solution casting, a transparent, elastic film was obtained, and the thermal behaviour of the material was studied. The melting region of the hard block in pTHFMD-U2, is much narrower than that of pTHF1100-U2. The tensile properties of both materials were compared, and illustrate a more pronounced strain hardening for the block copolymer with the monodisperse soft block.
107
Chapter 6
6.1 Introduction The chemical composition of a segmented copolymer is usually quite inhomogeneous. Both the hard and the soft block contain a distribution in segment length, which is inherent to the chemistry involved in their preparation. The block copoly(ether urea)s, which were described in the previous chapters, possess a uniform (monodisperse) hard block. However, the soft block of the polymer is still polydisperse, since the prepolymers of which it is made of, are polydisperse. It would be highly interesting to prepare and study such a segmented block copolymer having as well a monodisperse hard block as a monodisperse soft block. Some studies were preformed on the influence of the polydispersity of the soft block on the properties of thermoplastic elastomeric polyurethanes.1,2 Harrell and Cooper described the synthesis and properties of block copoly(ether urethane)s, in which the urethane part was based on piperazine (figure 6.1). These secondary urethane groups are unable of forming hydrogen bonds. According to the authors, this should facilitate the interpretation of structureproperties relationship. A prepolymer of low polydispersity (P.D.=1.1) was obtained, upon fractionation of a commercially available, polydisperse prepolymer. This was used as a monodisperse soft block, and built-in in the block copolymer. The material properties of this block copolymer were compared to those of an analogous block copolymer containing a polydisperse soft block (P.D.=1.7). The authors concluded that only some minor effects were observed upon lowering the polydispersity of the soft block: both strength and strain at break increased slightly.
O CN O O O NCO C4H8O y n NCO C4H8O CN x
Figure 6.1: Block copoly(ether urethane) possessing secondary urethane groups. Shirasaka et al., who studied block copoly(ether urethane urea)s possessing monodisperse soft blocks (P.D.1.1), found comparable results.1c Although this polymer showed a slightly lower strength and strain at break, the effects observed upon lowering the polydispersity of the soft block to approximately 1.1 were not significant. At this point the question arises, whether it would make sense to use soft blocks with even lower polydispersity, in other words, truly monodisperse soft blocks. Two examples are reported of segmented block copolymers possessing a uniform distribution in both hard and soft segment length.2 Such block copoly(ether ester)s were prepared by Wegner et al. (figure 6.2).2a,b They synthesized uniform butylene terephthalate-oligomers, as well as monodisperse pTHF-oligomers, via a stepwise synthesis. Both building blocks were coupled via a solution polycondensation to avoid transesterification, which would undo the uniformity of the hard 108
Block copolymers with a monodisperse soft block
blocks. The melting behaviour of the polymers was studied in detail. However, the limited amounts available of these polymers, due to the tedious purifications of the building blocks, hampered the study of their mechanical properties.
O C O O O CO C4H8O y n CO C4H8O C x
Figure 6.2: Block copoly(ether ester)s with uniform segments; x=14, y=12. Eisenbach and Baumgartner prepared block copoly(ether urethane)s comprising both uniform hard and soft segments (figure 6.3),2c using the same procedure as Wegner et al. The uniform block copoly(ether urethane)s were studied in detail with special emphasis on the thermal behaviour and hydrogen bonding. Again, no mechanical properties were discussed, unfortunately. They concluded that these well-defined block copolymers show better phase separation, and hence, sharper phase transitions.
O CN H CH2 O O CH2 H O NCO C4H8O y n NCO C4H8O CN x H H
Figure 6.3: Block copoly(ether urethane)s with uniform segments; x=14, y=12. In our opinion, it is highly intriguing to study the mechanical properties of a block copolymer containing a truly monodisperse soft block. We expect that such a material possesses unique tensile properties. An idealized representation of our idea is depicted in figure 6.4. On the left, a block copolymer with a polydisperse soft block is depicted. As the material is elongated, some of the shorter soft segments are already fully stretched, while some longer ones are not. As the stress becomes too high for the shorter chain, the material will fail, despite the fact that the longer chains can still bear more stress. On the right side of figure 6.4, the situation is depicted for a block copolymer with a monodisperse soft block. In this case, all chains are equally stretched as the material is elongated, and the stress is equally distributed over the chains. At the point were the chains become fully stretched, a disproportionate amount of force is required to elongate the material further, resulting in a steep upswing of the stress-strain curve. This strain hardening should lead to a much tougher material, compared to the block copolymer with a polydisperse soft block. Of course, this effect is only expected in highly oriented samples, in which the hard blocks are perfectly aligned as in figure 6.4. This requires very accurate processing.
109
Chapter 6
Polydisperse
Monodisperse
Elongation
hard block soft block
Figure 6.4: Schematic representation of a tensile experiment on a block copolymer, containing a polydisperse (left) and a monodisperse (right) soft block, respectively. Summering, our aim is to synthesize a block copoly(ether urea) having a uniform hard block and a monodisperse soft block, and to process this material into highly oriented, aligned samples. The procedure for the preparation of uniform hard blocks was described in section 3.4, but the procedure for the monodisperse soft block had to be developed. With a soft block length of approximately 1000 g/mol, typical for thermoplastic elastomers, this means that a monodisperse pTHF-oligomer comprising 13 monomer units, 1-(13), is needed. To perform mechanical testing, the material is needed on a multigram scale.
HO O H
13
1-(13)
6.2 Synthesis of monodisperse pTHF oligomers Poly(tetrahydrofuran), pTHF, is commercially available with molecular weights ranging from 650 4000 g/mol.3 Although it is synthesized by a living polymerization, the polydispersity of these prepolymers is rather high, approximately 1.7. The polydispersity can be lowered by fractionation. In principle, chromatographic techniques (SEC, HPLC) could be used to fractionate mixtures of oligomers, but in practice it is difficult to perform on a 110
preparative scale, since it needs an accurate process optimization, and is rather expensive.4 Fractionation of a mixture of pTHF-oligomers was successfully achieved by recrystallization. Especially the inclusion complexes of pTHF and urea were used for this purpose, and yielded fractions with molecular weights ranging from 800 2000 g/mol, and polydispersities from about 1.04.5 However, this molecular weight distribution is still too broad for our intentions. Therefore, we decided to develop a synthetic route for the desired monodisperse pTHFoligomers. Low molecular weight pTHF-oligomers have been prepared before by Wegner et al.6 The procedure they used, utilizes large quantities of LiAlH4 and AlCl3, and yielded a mixture of oligomers with D.P.1, which were separated by fractional distillation. However, for higher oligomers with D.P.>7, distillation was not possible, while these higher oligomers are the most interesting to be used. So, an easy and clean methodology was pursued which allowed the synthesis and purification of monodisperse oligomers of pTHF. The fastest way to obtain well-defined oligomers of high molecular weight is via a repetitive exponential growth algorithm, using two orthogonal protecting groups.7 Via this route, the chain length doubles after every reaction step. However, such an approach was unsuccessful in our case, due to non-quantitative reactions. The procedure that was investigated next, is based on a repetitive linear growth algorithm. Compared to the previous algorithm, the oligomeric chains grow less fast, since only one monomer unit per step is added per chain end. On the other hand, purification of the product is easier, because the low molecular weight (by)products are more easily separated from the higher molecular weight oligomers. Scheme 6.1 shows our synthetic sequence for the synthesis of monodisperse pTHFoligomers. First, benzyl 4-bromobutyl ether, 2, was synthesized according to a literature procedure.8 This reaction was performed on a relatively large scale of about 200 g. This monomer was coupled to a hydroxy-functionalized pTHF-oligomer 1-(x), i.e. trimer 1-(3), via the Williamson synthesis. This reaction is carried out in THF using sodium hydride as the base and 15-crown-5 to enhance the reactivity of the alkoxide. An excess of 2 was used to compensate for the amount that will be lost by elimination, and to increase the rate of the coupling reaction. After the reaction was complete, the reaction mixture contained besides the desired product also the excess of 2, and its elimination product. Subsequently, the benzyl ether groups were cleaved by hydrogenation in a Parr reactor at 60 psi hydrogen pressure. During this reaction, the double bond of the elimination product was hydrogenated as well. After hydrogenation, the hydroxy-terminated pTHF-oligomer containing two more monomeric units, 1-(x+2), was formed together with 1-butanol and 4-bromo-1-butanol. The latter was converted to THF by an intramolecular substitution reaction, upon treatment with sodium hydroxide; 1-butanol was removed by evaporation. Recrystallization of the product 111
Chapter 6
form diethyl ether yielded the monodisperse pTHF-oligomer as a white, crystalline powder. This product was used again as the starting compound for the next cycle. The overall yield of one cycle is about 80%.
OH
+
Br
Br
Bu4NHSO4 NaOH/H2O
O 2
Br
NaH, 15-crown-5 O 2 Br + HO 1-(x) O H x THF, 20C O 3-(x+2) + O H x+2 O + Br O x+2
HO 1-(x+2) + HO + NaOH
()
H2, Pd/C Br + CH3 EtOH
()
HO
()
Scheme 6.1: Synthetic cycle for the preparation of monodisperse pTHF-oligomers.
c
HO
d a d
b
O
12
OH
c
45.4 1.91 49.4
4.00
4.5
3.5
2.5
2 (ppm)
1.5
0.5
-0.5
Figure 6.5: 1H-NMR spectrum of pTHF-tridecamer, 1-(13).
112
After a twelve-step synthesis, the pTHF-oligomer with 13 units, 1-(13), was obtained on a scale of 20 g. All (intermediate) pTHF-oligomers 1-(x), x=3, 5, ..., 13 were successfully characterized by 1H-NMR, 13C-NMR, elemental analysis, SEC, and HPLC. The 1H-NMR spectrum of 1-(13) is shown in figure 6.5. The data are in good agreement with the structure proposed. The number average degree of polymerization based on this 1H-NMR spectrum is 12.4 (M=912 g/mol), which is within the experimental error of the expected value of 13. The SEC traces of the oligomers are shown in figure 6.6. An increase in hydrodynamic volume going from the trimer to the 13-mer is clearly observed. For all oligomers, the peaks in the chromatogram are very narrow, and polydispersities of 1.02 are calculated. However, this is the minimum limit that can be measured by this apparatus. Therefore, 1.02 is an upper limit for the polydispersity of the oligomer.
0,012
x=13
x=3
0,010
0,008
A.U.
0,006 Volts
0,004
0,002
0,000
-0,002
-0,004
10
Elution time (min)
Figure 6.6: SEC traces of pTHF-oligomers 1-(x), x = 3, 5, , 13. MALDI-TOF MS and HPLC were used to obtain a more accurate value for the polydispersity, since these techniques have a better resolution for the identification of the different oligomers. In figure 6.7, the MALDI-TOF spectrum of 1-(13) is shown. The peak corresponding to the sodium adduct of the 13-mer (m/z=977) is observed, as well as the potassium adduct. However, in addition to the 13-mer, also some lower oligomers were observed. The presence of these oligomers was confirmed by HPLC. Based on the results found with MALDI-TOF MS and HPLC, an estimation of the composition of the oligomermixture was made, from which a polydispersity of 1.004 was calculated. This value is very low, and confirms the monodispersity of the prepared prepolymer. Currently, we attempt to improve the synthetic procedure, by monitoring the extent of the reaction more quantitatively by HPLC9, in order to lower the polydispersity even further. 113
Chapter 6
(12 )+Na+ 905.72 (13 )+Na+ 977.82
Intensity
(11 )+Na+ 833.63
(10 )+Na+ 761.52 849.60
922.71
(13 )+K+ 993.80
500
700
900
1100
1300
1500
mass (m/z)
Figure 6.7: MALDI-TOF spectrum of pTHF-tridecamer, 1-(13).
6.3 Synthesis of a block copolymer containing a monodisperse soft block The monodisperse pTHF-oligomer was used for the synthesis of a block copoly(ether urea) having a uniform hard block and a monodisperse soft block. The synthesis is depicted in scheme 6.2, and is comparable to the synthesis of the block copoly(ether urea)s possessing a polydisperse soft block, which was described in chapter 3. First the hydroxy-terminated pTHF-oligomer, 1-(13), was converted to an amine-terminated prepolymer, 4, via cyanoethylation and reduction. The next step involved the chain extension of 4 with 1,4butanediisocyanate in chloroform at room temperature, to give the block copolymer with a monodisperse soft block and two urea groups in the hard block, pTHFMD-U2. Characterization of this polymer by FT-IR showed the presence of strong hydrogen bonds and the FT-IR spectrum is identical to the spectrum of its polydisperse analogue pTHF1100-U2. Also the 1H-NMR spectrum is in agreement with the proposed structure. The molecular weight distribution of the polymer was determined by SEC with NMP as the eluent: Mn=23*103 g/mol and Mw/Mn=1.8.
114

CN +
HO 1-(13) BH3 THF
H
13
base
NC
CN
13
H2N
O 4
O
13
NH2
OCN
NCO
O O O
13
H N H N O
H N n
N H
pTHFMD-U2
Scheme 6.2: Synthesis of block copoly(ether urea) pTHFMD-U2. During the synthesis of pTHFMD-U2, a remarkable phenomenon occurred. The reaction mixture formed a completely transparent and dry gel. Gelling of compounds containing two urea groups is often observed, due to their strong intermolecular hydrogen bonding.10 Nevertheless, it was not observed for the block copoly(ether urea) with the polydisperse soft block, pTHF1100-U2, which is synthesized under exactly the same conditions. Apparently, the transition from a polydisperse to a monodisperse soft block, leads to more order in the system, facilitating the stacking of the urea groups. The gelation does not occur when dimethylacetamide is used as the solvent, which illustrates that it is not caused by covalent crosslinking of the polymer. In order to prepare films, suitable for mechanical characterization, the gelled material was placed in a Petri-dish, and covered to prevent evaporation of the solvent. After a few days, the gel had flowed over the entire bottom of the dish, and after evaporation of the solvent, a homogeneous, transparent, elastic film was formed, out of which tensile bars could be punched.
6.4 Properties Thermal properties The thermal behaviour of block copoly(ether urea) pTHFMD-U2 was determined by optical microscopy. A flow temperature of 125C was found, which is slightly lower than the flow temperature of the material with the polydisperse soft block. The flow temperature was confirmed by DSC (figure 6.8 and table 6.1). The melting and crystallization temperature of the polymer with the monodisperse soft block are slightly lower than for those with the 115
Chapter 6
polydisperse soft block. The melting enthalpy of both polymers is equal, which is not that surprising, since both contain roughly the same content of hard block, due to the same average soft block length. The melting peak of pTHFMD-U2 is more narrow than that of pTHF1100-U2. This is in agreement with previous studies, that describe a positive correlation between the polydispersity of the soft block, and the broadness of this phase transition.11 So, the better defined the block copolymer, the narrower its melting region. The narrower phase transition was also observed in real-time SAXS/WAXD/DSC measurements of pTHFMD-U2. Preliminary results reveal a sharper drop of the total scattering power as a function of temperature, compared to the less well-defined pTHF1100-U2 (section 4.4). No crystallization of the monodisperse soft block in the block copolymer is observed. Apparently, it is too short to crystallize within its restricted environment. The subzero part of the DSC is rather unclear, and no neat glass transition is observed. However, at a heating rate of 40C, and starting the trace at 150C, a glass transition temperature at 74C was noticed.
27
Heat Flow Endo Up (mW)
25
23 Tm,h 21 Tg
19 Tc,h 17 -100 -50 0 50 T (C) 100 150 200
Figure 6.8: DSC traces of pTHFMD-U2 (top), and pTHF1100-U2 (bottom) at a heating/ cooling rate of 10C/min, second heating runs are shown. Table 6.1: Thermal transitions of block copoly(ether urea)s as determined by DSC; Subscript s denotes the soft block, h denotes the hard block. Copolymer pTHFMD-U2 pTHF1100-U2
a) Determined at a heating rate of 40C/min.
Tga (C) 74 68
Tm,h (C) 122 131
Hm,h (J/g) 14.4 14.2
Tc,h (C) 94 102
116
Tensile properties Tensile bars of pTHFMD-U2 were obtained from non-oriented films after solutioncasting. In figure 6.9, its stressstrain curve is compared to that of its polydisperse analogue. The Youngs modulus is slightly lower for the monodisperse one, E=53 MPa versus 96 MPa; its yield stress is also lower, yield=6.5 MPa versus 9.6 MPa. The curve of pTHFMD-U2 shows a moderate strain softening effect, followed by a pronounced strain hardening. This steep upswing toughens the material. This effect is not caused by strain induced crystallization, but seems to be a direct result of the monodispersity of the soft block in the block copoly(ether urea), possibly caused by the effect described earlier (figure 6.4). The strain at break of this polymer is lower than that of pTHF1100-U2, break=790% versus 1060%. This can be an intrinsic property of the monodisperse polymer, or it can be due to the relatively poor quality of the film, that was obtained from a gelled solution. Although the tensile properties of both materials do not differ that much, we expect a more spectacular effect of the monodisperse soft block upon orientating the sample, and aligning the hard blocks.
40
30
pTHFMD-U2
pTHF1100-U2
(MPa)
20
10
0 0 500 (%) 1000 1500
Figure 6.9: Stressstrain curves of non-oriented block copoly(ether urea)s with polydisperse (pTHF1100-U2) and monodisperse soft block (pTHFMD-U2), respectively.
117
Chapter 6
6.5 Conclusions and outlook The stepwise synthesis of monodisperse pTHF-oligomers, 1-(x), was successfully performed. The advantage of this synthetic procedure is the ability to use an excess of reagents, to ensure completion of the coupling reaction, and the facile removal of starting- and byproducts. After twelve steps, a monodisperse pTHF 13-mer, 1-(13), was synthesized on a scale of 20 gram. Extensive analysis of this oligomer by 1H-, 13C-NMR, FT-IR, SEC, HPLC, and MALDI-TOF MS, successfully proved the proposed structure, and revealed a molecular weight of 912, and a polydispersity of 1.004. This means that a conversion of 95% per step per end group is achieved. Currently, we are optimizing the synthetic procedure and characterization, and trying to reduce the polydispersity even more. The hydroxy-end groups of this oligomer were converted to amine-end groups, via a two step procedure, which was previously discussed in chapter 3. This prepolymer was chain extended by reaction with 1,4-butanediisocyanate, to give a block copoly(ether urea), pTHFMD-U2, with a uniform hard block, and a monodisperse soft block. During this synthesis, the solution formed a gel due to hydrogen bonding between the urea groups. This was surprising, since the analogous polymer with the polydisperse soft block, pTHF1100-U2, does not gel. Apparently, the better defined soft block results in more order in the system, facilitating the stacking of the urea groups. The reversibility of the gel formation indicated the absence of covalent crosslinking. A transparent, elastic film was obtained upon solution casting of the polymer. Thermal analysis of this material by optical microscopy and DSC, showed the reversible melting of the hard block at about 125C, causing the material to flow. This temperature is slightly lower than that of pTHF1100-U2, and the melting region of the polymer with the monodisperse block is much narrower, due to the better defined soft block. This behaviour is in agreement with previous studies, that indicate sharper phase transitions for (block co)polymers of low polydispersity.11 Tensile testing of the non-oriented polymer pTHFMD-U2, illustrated more strain hardening, and a slightly lower Youngs modulus, yield stress and strain at break compared to pTHF1100-U2. The same trend, although less pronounced, was observed by Shirasaka et al. for TPUs containing soft blocks of low polydispersity (P.D.=1.1).1c Additional experiments are necessary to get more insight into the effect of the monodisperse soft block on the material properties. In summary, we successfully prepared a thermoplastic elastomeric block copolymer possessing both monodisperse soft and uniform hard segments, in which the hard blocks associate by well-controlled supramolecular interactions. Accordingly, its molecular structure resembles that of spider dragline silk, Natures high-performance fiber. We are now at a point 118
where we can optimize the processing of the material to obtain highly oriented fibers, e.g. by gel spinning.12 Subsequently, we can study the mechanical properties of the material and validate the model that is depicted in figure 6.4, and which predicts a very pronounced effect of the monodispersity of the soft block on the tensile properties.

General methods and instrumentation. General methods concerning purification of solvents and spectroscopic techniques can be found in the experimental section of previous chapters. Gas chromatography-mass spectrometry (GC-MS) was performed on a Shimadzu GC-17A gas chromatograph equipped with a Shimadzu GCMS-QP5000 gas chromatograph mass spectrometer. Temperature program: 1 min at 50C, heat at 40C/min to 300C, wait for 5 min. SEC on the pTHF-oligomers was performed on the same system as described in chapter 2, but with chloroform as the eluent. HPLC was performed on benzoyl derivatives of hydroxy-terminated pTHF-oligomers, employing a Shimadzu Low pressure gradient HPLC system, using a LC10-AT pump, a SPD-10AV UV-Vis detector, and a Alltima C18 5 (150*3.2 mm) column. A gradient in eluent was used, starting with pure acetonitrile for 7 min, then adding THF in a rate of 1%/min. Materials. The availability of some chemicals was discussed in the experimental section of previous chapters. Sodium hydride (60% dispersion in mineral oil), tetrabutylammonium hydrogensulfate, and 15-crown-5 were purchased from Aldrich; benzyl alcohol from BDH; 1,4-dibromobutane, and 1,4-butanediol from Merck. Benzyl 4-bromobutyl ether, 2 Sodium hydroxide (100 g, 2.5 mol) was slowly dissolved in water (400 ml) and cooled to 20C. Benzyl alcohol (107.5 g, 1.00 mol), 1,4-dibromobutane (500 g, 2.3 mol) and tetrabutylammonium hydrogensulfate (8.5 g, 25 mmol) were added, and the reaction mixture was stirred at 70C for 4 h. Water (500 ml) was added and the product was extracted with hexane (500 ml). The organic layer was dried with magnesium sulfate, filtered, and concentrated. The excess of dibromobutane was recovered by distillation (29C/0.35 mbar). The product was distilled two times (72C/1.5*10-2 mbar) and obtained as a colourless liquid (209 g, 86%). The product is stored under argon, in the dark at 4C to prevent it from oxidation. 1H-NMR (CDCl3): 7.4-7.2 (m, 5H, Ph-H), 4.50 (s, 2H, PhCH2), 3.50 (tr, 2H, OCH2CH2, J=6.2 Hz), 3.43 (tr, 2H, BrCH2, J=6.6 Hz), 1.97 (m, 2H, OCH2CH2), 1.76 (m, 2H, BrCH2CH2). 13C-NMR (CDCl3, 100 MHz): 138.4, 72.9, 69.2, 33.7, 29.6, 29.3. FT-IR (ATR): 2856, 1496, 1454, 1360, 1248, 1101, 733, 696 cm-1. GC-MS: tr=4.27 min, <1% 1,4-dibromobutane, <1% benzyl bromide. 5,10-Dioxa-tetradecane-1,14-diol, pTHF 3-mer, 1-(3) Sodium (5.75 g, 0.25 mol) was cut in small pieces, washed with heptane (100 ml), and reacted under dissolution in 1,4-butanediol (45.0 g, 0.50 mol) at 70C while stirring vigorously. The reaction mixture was heated to 100C and 1,4-dibromobutane (21.6 g, 0.10 mol) was added dropwise. The mixture was stirred from 3 h, after which it was cooled to room temperature. Water (40 ml) was added, and the aqueous layer was extracted four times with dichloromethane (70 ml). The combined organic layers were extracted one more time with water (20 ml), and was then dried with magnesium sulfate, and concentrated in vacuo. The yellow product was dissolved in ether (250 ml), and recrystallized at 18C overnight. The ether was decanted, the product washed with cold ether (100 ml), and redissolved in ether (100 ml). The ether was evaporated in vacuo on a rotary evaporator without putting the flask in the water bath. This allowed the product to precipitate during drying, and yielded a fine, white powder (7.02 g, 30%). Tm=36C (Lit: 33C). GC-MS: tr=5.08 min, >98 % pure. 1H-NMR (CDCl3): 3.64
119
Chapter 6
(q, 4H, CH2OH, J=5.9 Hz), 3.46 (t, 8H. CH2OCH2, J=5.9 Hz), 2.62 (t, 2H, OH, J=5.9 Hz), 1.68 (m, 12H, OCH2CH2CH2CH2O). 13C-NMR (CDCl3): 70.8 (CH2CH2CH2CH2OH), 70.7 (OCH2CH2CH2CH2O), 62.7 (CH2OH), 30.4 (CH2CH2OH), 26.9 (CH2CH2CH2OH), 26.3 (OCH2CH2CH2CH2O). Anal. Calcd. (%) for C12H26O4: C 61.51, H 11.18. Found (%): C 61.31, H 11.33. FT-IR (ATR): 3361, 2937, 2859, 1446, 1370, 1105, 1057 cm-1. MALDI-TOF [M+Na+] = Calcd. 257.2 Da. Obsd. 256.8 Da. oligo-pTHF, 1-(x) The synthesis of pTHF 5-mer, 1-(5) is given as a typical example. NaH (15.93 g 60% disp., 0.40 mol) was washed two times with pentane (100 ml), and decanted carefully. THF (200 ml) was added to this, and the suspension was cooled on an ice bath. Diol, 1-(3), (11.70 g, 0.50 mol) was dissolved in THF (50 ml), and added dropwise to the suspension. Be careful: hydrogen gas evolution! After this, 15-crown-5 (4.40 g, 0.020 mol) was added. The suspension was stirred at room temperature for 30 min under an argon atmosphere. Subsequently, benzyl 4-bromobutyl ether, 2, (48.60 g, 0.20 mol) was added dropwise. The reaction mixture was stirred for 4 h at room temperature. After the reaction was complete, the reaction mixture was cooled on an ice bath, and water (20 ml) was added slowly. Be careful: hydrogen gas evolution! When evolution of gas ceased, the reaction mixture was evaporated to dryness, and sodium hydroxide solution (100 ml 1 M) was added, and this was extracted two times with ether (200 ml). The combined organic layers were concentrated, and redissolved in ethanol (100 ml). Water (2 ml), sulfuric acid (0.5 g), and palladium (0.1 g 10% on carbon) were added to this. The reaction mixture was flushed with nitrogen. Subsequently, it was hydrogenated at 65 psi and room temperature in a Parr reactor for several hours, until the pressure no longer dropped. The reaction mixture was flushed with nitrogen, filtered, neutralized with sodium hydroxide solution and concentrated. Then it was poured in sodium hydroxide solution (70 ml 1M) and stirred for 15 min, after which it was extracted three times with ether (100 ml). The combined ether layers were dried with sodium sulfate, filtered, and ether was evaporated under reduced pressure. The residue was recrystallized from ether (80 ml) at 18C. The white powder was filtered off, washed with cold ether and dried in vacuo (14.87 g, 78 %). Tm=41C (Lit: 39C). GC-MS: tr=7.2 min, >98 % pure. 1H-NMR (CDCl3): 3.63 (q, 4H, CH2OH, J=5.9 Hz), 3.473.40 (m, 16H. CH2OCH2), 2.62 (t, 2H, OH, J=5.7 Hz), 1.691.62 (m, 20H, OCH2CH2CH2CH2O). 13CNMR (CDCl3): 70.8, 70.6, 70.5, 62.7, 30.4, 26.9, 26.5, 26.4, 26.4. Anal. Calcd. (%) for C20H42O6: C 63.46, H 11.18. Found (%): C 63.14, H 11.39. MALDI-TOF [M+Na+] = Calcd. 401.3 Da. Obsd. 401.0 Da. pTHF 7-mer, 1-(7) Yield: 24.5 g (74%). Tm=36C (Lit: 36C). 1H-NMR (CDCl3): 3.64 (q, 4H, CH2OH, J=5.4 Hz), 3.473.42 (m, 24H. CH2OCH2), 2.56 (t, 2H, OH, J=5.6 Hz), 1.701.62 (m, 28H, OCH2CH2CH2CH2O). Anal. Calcd. (%) for C28H58O8: C 64.33, H 11.18. Found (%): C 64.11, H 11.27. MALDI-TOF [M+Na+] = Calcd. 545.4 Da. Obsd. 545.2, 473.2 Da. pTHF 9-mer, 1-(9) Yield: 29.35 g (78%). Tm=31C. 1H-NMR (CDCl3): 3.64 (q, 4H, CH2OH, J=5.9 Hz), 3.483.41 (m, 32H. CH2OCH2), 2.51 (t, 2H, OH), 1.691.60 (m, 36H, OCH2CH2CH2CH2O). 13C-NMR (CDCl3): 70.8, 70.6, 70.5, 62.7, 30.4, 26.9, 26.5, 26.4, 26.4. Anal. Calcd. (%) for C36H74O10: C 64.83, H 11.18. Found (%): C 64.47, H 11.18. MALDI-TOF [M+Na+] = Calcd. 689.5 Da. Obsd. 689.4, 617.3 Da. pTHF 11-mer, 1-(11) Yield: 21.9 g (61%). Tm=32C. 1H-NMR (CDCl3): 3.64 (4H, CH2OH), 3.473.40 (40H. CH2OCH2), 2.52 (2H, OH), 1.671.62 (44H, OCH2CH2CH2CH2O). 13C-NMR (CDCl3): 70.8, 70.6, 70.5, 62.7, 30.3, 26.9, 26.5. Anal. Calcd. (%) for C44H90O12: C 65.15, H 11.18. Found (%): C 65.27, H 11.16. MALDI-TOF [M+Na+] = Calcd. 833.6 Da. Obsd. 833.4, 761.3, 689.2 Da. pTHF 13-mer, 1-(13) Yield: 20.0 g (77%). Tm=31C. 1H-NMR (CDCl3): 3.63 (q, 4H, CH2OH, J=5.9 Hz), 3.473.40 (m, 46H. CH2OCH2), 2.61 (t, 2H, OH, J=5.7 Hz), 1.701.60 (m, 50H, OCH2CH2CH2CH2O). 13C-NMR (CDCl3): 70.8,
120
70.6, 70.5, 62.6, 30.3, 26.9, 26.5, 26.4, 26.4. Anal. Calcd. (%) for C52H106O14: C 65.37, H 11.18. Found (%): C 65.18, H 11.35. MALDI-TOF [M+Na+] = Calcd. 977.8 Da. Obsd. 977.8, 905.7, 833.6, 761.5 Da. Bis-(3-aminopropyl)-pTHF 13-mer, 4 Synthetic procedure as described in chapter 3 for bis(3-aminopropyl)-poly(tetrahydrofuran). 1 H-NMR (CDCl3): 3.49 (t, 4H, OCH2CH2CH2NH2, J=6.2), 3.41 (m, 60H, OCH2CH2CH2CH2O), 2.79 (t, 4H, CH2NH2, J=7.6), 1.71 (t, 4H, OCH2CH2CH2NH2, J=6.6), 1.62 (m, 56H, OCH2CH2CH2CH2O), 1.12 (s, 4H, NH2). 13 C-NMR (CDCl3): 70.6 (OCH2CH2CH2CH2O), 68.9 (OCH2CH2CH2NH2), 39.7 (CH2NH2), 33.6 (OCH2CH2CH2NH2), 26.5 (OCH2CH2CH2CH2O). FT-IR (ATR): 3361, 2942, 2863, 1492, 1371, 1106, 995 cm1 . pTHFMD-U2 Bis-(3-aminopropyl)-pTHF 13-mer ( 0.86 g, 0.81 mmol) was dissolved in chloroform (8 ml), and 1,4butanediisocyanate (0.113, 0.81 mmol) in chloroform (2 ml) was added dropwise. Within 30 min the reaction mixture gelled, after which it was poured in a Petri-dish, and allowed to flow for 7 days without evaporation of the solvent. Then, it was dried in vacuo to yield a completely transparent, elastic film (0.93 g, 96%). 1H-NMR (DMSO, 50C): 5.69 (4H, NH), 3.34 (57H, CH2O + H2O), 3.00 (8H, CH2N), 1.51 (52H, OCH2CH2CH2N + OCH2CH2CH2CH2O), 1.33 (4H, NCH2CH2CH2CH2N). FT-IR (ATR): 3327 (NH stretching), 2940, 2855, 1617 (C=O stretching), 1578, 1367, 1105 (CO stretching) cm-1. SEC (NMP, rel. to PS): Mn=23*103 g/mol.

1. a) Harrell, L. L. Macromolecules 1969, 2, 607; b) Ng, H. N., Allegrezza, A. E., Seymour, R. W. and Cooper, S. L. Polymer 1973, 14, 255; c) Shirasaka, H., Inoue, S.-i., Asai, K. and Okamoto, H. Macromolecules 2000, 33, 2776. 2. a) Schmidt, H. G. Ph.D. Thesis, University of Freiburg (Freiburg), 1984; b) Wegner, G., in Legge, N. R., Holden, G. and Schroeder, H. E. Thermoplastic elastomers: A comprehensive review; Carl Hansser Verslag: New York, 1987, Section 12/5; c) Eisenbach, C. D., Baumgartner, M. and Gunter, G. Adv. Elastomers Rubber Elasticity, [Proc. Symp.] 1985, 51. 3. Poly(tetrahydrofuran) is sold by e.g. Sigma-Aldrich Co. under the trade name Terathane polyether glycol. 4. More information about monodisperse poly(ethylene glycol)-oligomers obtained by low-pressure chromatography, can be found at www.Polypure.no. 5. Schmidt, G., Enkelmann, V., Westphal, U., Droescher, M. and Wegner, G. Colloid Polym. Sci. 1985, 263, 120. 6. Bill, R., Droescher, M. and Wegner, G. Makromol. Chem. 1981, 182, 1033. 7. Percec, V. and Asandei, A. D. Macromolecules 1997, 30, 7701. 8. Comins, D. L., LaMunyon, D. H. and Chen, X. J. Org. Chem. 1997, 62, 8182. 9. Andrews, G. D., Vatvars, A. and Pruckmayr, G. Macromolecules 1982, 15, 1580. 10. a) Hanabusa, K., Shimura, K., Hirose, K., Kimura, M. and Shirai, H. Chem. Lett. 1996, 885; b) van Esch, J., De Feyter, S., Kellogg, R. M., De Schryver, F. and Feringa, B. L. Chem. Eur. J. 1997, 3, 1238; c) Loos, M. d., Esch, J. v., Stokroos, I., Kellogg, R. M. and Feringa, B. L. J. Am. Chem. Soc. 1997, 119, 12675; d) van Esch, J., Kellogg, R. M. and Feringa, B. L. Tetrahedron Lett. 1997, 38, 281; e) Van Esch, J., Schoonbeek, F., De Loos, M., Kooijman, H., Spek, A. L., Kellogg, R. M. and Feringa, B. L. Chem. Eur. J. 1999, 5, 937; f) Schoonbeek, F. S., Van Esch, J. H., Wegewijs, B., Rep, D. B. A., De Haas, M. P., Klapwijk, T. M., Kellogg, R. M. and Feringa, B. L. Angew. Chem. Int. Ed. 1999, 38, 1393; g) Schoonbeek, F. S., Van Esch, J. H., Hulst, R., Kellogg, R. M. and Feringa, B. L. Chem. Eur. J. 2000, 6, 2633; h) Brinksma, J., Feringa, B. L., Kellogg, R. M., Vreeker, R. and van Esch, J. Langmuir 2000, 16, 9249; i) De Loos, M., Ligtenbarg, A. G. J., Van Esch, J., Kooijman, H., Spek, A. L., Hage, R., Kellogg, R. M. and Feringa, B. L. Eur. J. Org. Chem. 2000, 3675; j) De Loos, M., Van Esch, J., Kellogg, R. M. and Feringa, B. L. Angew. Chem. Int. Ed. 2001, 40, 613; k) Estroff, L. A. and Hamilton, A. D. Angew. Chem. Int. Ed. 2000, 39, 3447; l) Boileau, S., Bouteiller, L., Laupretre, F. and Lortie, F. New J. Chem. 2000, 24, 845; m) Lortie, F., Boileau, S., Bouteiller, L., Chassenieux, C., Deme, B., Ducouret, G., Jalabert, M., Laupretre, F. and Terech, P. Langmuir 2002, 18, 7218.
121
Chapter 6
11. a) Angerman, H., ten Brinke, G. and Erukhimovich, I. Macromolecules 1998, 31, 1958; b) Kane, L., Satkowski, M. M., Smith, S. D. and Spontak, R. J. J. Polym. Sci. B 1997, 35, 2653; c) Sakurai, S. and Nomura, S. Polymer 1997, 38, 4103. 12. a) Smith, P. and Lemstra, P. J. J. Mater. Sci. 1980, 15, 505; b) Ward, I. M. Structure and properties of oriented polymers; 2 ed.; Chapman & Hall: London, 1997.
122
7
Reversible networks based on dendrimer telechelic association
Abstract Association of a dendritic host possessing 32 binding sites, and a telechelic polymer containing two guest groups, results in the formation of a reversible network. The structure of these complexes in solution depends strongly upon the concentration. At low concentration, the two end groups of one bifunctional guest are proposed to complex to the same host, and flower-like structures are formed. At higher concentrations, both end groups of one guest are complexed to different hosts, forming a bridge between them. This gives rise to the formation of larger associates, and eventually a transient network. This process was confirmed by both viscometry and dynamic light scattering. The viscosity of a solution of the complex is strongly dependent on the concentration and the temperature. A monofunctional, polymeric guest not capable of forming a network structure was used as a reference. A bifunctional guest with a monodisperse molecular weight distribution was synthesized to study the influence of the polydispersity on the transition from flowers to a network. Whereas a polydisperse, bifunctional guest gives rise to a gradual transition, the monodisperse, bifunctional guest shows a sharp transition, which results in a two phase system: a dilute, low-viscosity layer containing exclusively flowers, and a concentrated, highly viscous layer which contains a network.
123
Chapter 7
7.1 Introduction The class of thermoplastic elastomeric block copolymers can roughly be divided into segmented copolymers and ABA-triblock copolymers.1 Examples of the first type of TPEs have been described extensively in the previous chapters. In this chapter, we would like to discuss a system that shows strong resemblance with the ABA-triblock copolymers. The main groups of thermoplastic elastomeric triblock copolymers consists of the polystyrene/polybutadiene (SBS) triblock copolymers.2 The molecular weights of both blocks are approximately 10 15 and 30 70 kg/mol, respectively. The incompatibility and the high molecular weight of the blocks cause the system to microphase separate, resulting in the formation of glassy domains of polystyrene (high Tg) that are embedded in the amorphous polybutadiene phase (low Tg) (figure 7.1). Due to the mechanical stability of these glassy domains, they serve as reversible crosslinks that hold together the rubbery polybutadiene phase, causing this material to be elastomeric. The mechanical stability of the polystyrene domains is lost above its glass transition temperature, allowing the material to flow like a typical thermoplastic polymer.
Polystyrene Polybutadiene Polystyrene
Figure 7.1: Schematic structure and morphology of a SBS-triblock copolymer. Dissolution of the SBS-triblock copolymer in a solvent that selectively dissolves the middle polybutadiene block, leads to micellization of the polystyrene blocks.3 At high concentrations, both polystyrene blocks of one and the same macromolecule are part of different micelles, and the polybutadiene block forms a bridge between neighboring micelles, giving rise to the formation of a reversible network (gel). On the other hand, under dilute conditions, the two polystyrene blocks of one polymer chain are aggregated within one micelle, forcing the polybutadiene block to make a loop, so that a flower-like structure is formed.
124
Systems that show very similar behaviour include the hydrophobically end-capped urethane-coupled poly(ethylene oxide)s (HEURs)4 and telechelic ionomers5. Especially the first system has been studied extensively.6 Water is used as the selective solvent for the PEO block, forcing the hydrophobic chain ends to aggregate into micelles. This system shows highly interesting rheological properties7, and is applied as associative thickener in coatings4b,6a, as sieving media for DNA sequencing8, or as gel for controlled drug release9. The association of the chain ends in these ABA-triblock copolymers is ill-defined, since the phase separation causing the micellization has poor directionality and specificity. This limits the tuning of important parameters that influence the properties of these reversible networks, like the functionality of the reversible crosslinks (micelles). In supramolecular polymers, in which the monomeric units are held together by reversible secondary interactions (e.g. hydrogen bonding, electrostatic interactions or metalligand complexation), the association of molecules or functional groups is highly directional and specific.10 Supramolecular polymers hold promise as a unique class of novel materials, because they combine many of the attractive features of conventional polymers (e.g. mechanical strength) with properties that result from the reversibility of the bonds between the monomer units. This results in materials that are able to respond to external stimuli (smart materials), and that are always in the thermodynamically most stable state. In contrast to supramolecular (homo)polymers, the field of supramolecular (multi)block copolymers is rather unexplored.11 In these systems, the individual blocks are linked via their termini by non-covalent interactions. This principle can give rise to novel, interesting materials, since combinations of blocks can be made which were previously inaccessible due to the absence of a suitable polymerization technique, and since the reversibility of bonds makes the material responsive to external stimuli. Supramolecular block copolymers possessing reversible crosslinks based on highly directional and specific supramolecular interactions are unknown until now, although they may yield reversible networks with highly interesting properties. Recently, Maurice Baars and other colleagues in our laboratory studied the reversible binding of guest molecules to a multivalent dendritic host. The guests contain a ureido-acetic acid unit that complexes to an adamantylurea functionalized poly(propylene imine) dendrimer via strong and directional, multiple interactions: both an electrostatic interaction between the protonated tertiary amine and the carboxylate moiety, and hydrogen bonding between the urea groups of host and guest are responsible for this (scheme 7.1).12 The complexation was studied with FT-IR, 1H-NMR, NMR relaxation, NOESY NMR, and isothermal microcalorimetry, and association constants in chloroform ranging from 104 105 l/mol were determined, depending on the nature of the guest. A maximum of 32 guest molecules are bound to a fifth generation dendritic host, bearing 64 adamantylurea end groups. 125
Chapter 7
H N O N 32 H N Host O + O HO Guest H N O
H N H N O + NH O O H N O H N H N R O Complex H N H N R H N
H N 32
Scheme 7.1: Proposed scheme for the complexation of a dendritic host with a ureido-acetic acid modified guest molecule. This raised the question, what would happen when these multivalent hosts were complexed with the end groups of bifunctional polymers (telechelics). In principle, this is analogous to the polymerization of two multifunctional monomers, and hence, it should result in a network structure.13 Our hypothesis is that in dilute solutions, intramolecular complexation is favoured over intermolecular complexation, giving rise to the formation of loops (scheme 7.2 left). These associates are referred to as flower-like structures. At high concentrations, intermolecular complexation is favoured over intramolecular complexation, and the bifunctional guest forms a bridge between two dendritic hosts, resulting in a transient network (scheme 7.2 right). In the intermediate concentration-region, large associates are present in which several dendritic hosts are held together by the bifunctional guests (scheme 7.2 middle). Since our system is reversible, and therefore dynamical, the transition from the flower-like structures to a network is spontaneous and might give rise to unique rheological properties. In this chapter, the synthesis of several polymeric guests is described, together with a study of their complexation with the dendritic host. Some rheological properties of these complexes in solution at different concentrations are discussed.
126
dilute semi-dilute concentrated Scheme 7.2: Schematic representation of the transition from flowers to a network.
7.2 Synthesis of host and guests As mentioned above, the multivalent, dendritic host is an adamantylurea functionalized poly(propylene imine) dendrimer of the fifth generation, and is denoted as DAB-dendr-(NHCONH-Ad)64. Its third generation analogue is depicted below. The fifth generation host possesses 64 adamantylurea end groups, and hence 32 binding sites. These adamantylurea functionalized dendrimers were synthesized previously, by reaction of an amine-terminated poly(propylene imine) dendrimer and 1-adamantyl isocyanate.12a,25 These hosts were successfully characterized by 1H-, 13C-NMR, FT-IR, and FAB-MS.
NH NH O NH O O NH
NH O NH
NH NH O
N NH O N NH O H N O H N O NH N H N N N H N N N H N
N H
N H O
N H O
N H
N H N N N H O
N H
N H
N O
NH NH O
HN NH
HN O HN
HN O NH
DAB-dendr-(NHCONH-Ad)16
127
Chapter 7
Both monofunctional, 1-(x), and bifunctional, 2-(x), polymeric guests were prepared (scheme 7.3 and 7.4). The number between parentheses denotes the number-averaged molecular weight of the guest, as determined by 1H-NMR. Their backbone consists of a pTHF prepolymer, to ensure good solubility in a range of solvents. The monofunctional guest is used as a reference, since it is expected to be unable to form a network. Monofunctional pTHF was prepared by a living cationic ring-opening polymerization of THF.14 Via a procedure comparable to the one used for the synthesis of heterofunctional pTHF (section 3.4), -aminobutyl--methyl-pTHF was prepared by initiation with methyl trifluoromethanesulfonate and termination with hexamethylenetetraamine (scheme 7.3). The molecular weight of the product is confined by the conversion at which the polymerization is terminated. A prepolymer with a number-averaged molecular weight of 1150 g/mol was obtained. The amine end group of the polymer was reacted with ethyl isocyanatoacetate. Subsequent hydrolysis of the ester functionality and acidification yielded the monofunctional guest 1-(1650). The reaction mixture should by acidified mildly and extracted immediately, to prevent the formation of hydantoin. The product was well soluble in chloroform, and successfully characterized by 1H-NMR and FT-IR. Due to some fractionation of the polymer during work-up, the number-averaged molecular weight increased slightly to 1650 g/mol.
O O S CF3 O O n-1 + + n O DTBP HMTA 1) H 2SO4 / MeOH 2) base O NH2 O O O n-1 N H O O O n-1 N H 1-(x) N H O OH N H O O NaOH THF O C N O CHCl3
Scheme 7.3: Synthesis of monofunctional polymeric guest. Telechelic polymers bearing two ureido-acetic acid functionalities, 2-(x), were derived from amine-terminated pTHF-prepolymers that were described in chapter 3. Just as the monofunctional prepolymer, they were reacted with ethyl isocyanatoacetate, and hydrolyzed (scheme 7.4). In this way, a series of bifunctional guests, 2-(x), with different molecular weights was obtained (table 7.1). Again, due to several extractions during work-up, fractionation of the polymer leads to higher molecular weights than expected based on that of 128
the starting prepolymer. We failed to synthesize shorter bifunctional guests, because these are insoluble in chloroform, which hampered their purification. During the work-up of these shorter guests, considerable amounts of hydantoin were formed.
H2 N O O HO O N H N H 2-(x) O O n O n H N O H N O OH NH2
Scheme 7.4: Synthesis of bifunctional, polymeric guests. Table 7.1: Bifunctional, polymeric guests. Prepolymer 2-(1550) 2-(4500) 2-(1270, MD) Mna (g mol1) 1550 4500 1270 Mw/Mnb 1.7 1.5 1.003
a) As determined by 1H-NMR. b) Based on the polydispersity of the amine-terminated prepolymer.
The last entry in table 7.1, 2-(1270, MD), is a bifunctional guest, based on the monodisperse pTHF-oligomer that was described in chapter 6. Its molecular weight is comparable to that of 2-(1550), but its polydispersity is much lower. The influence of this much narrower molecular weight distribution on the complexation, and the transition from flowers to a network is studied in the following sections.
7.3 Complexation of polymeric guests and dendritic host Several techniques have been used to study the complexation of guests containing the ureido-acetic acid unit with an adamantylurea functionalized dendrimer of the fifth generation, e.g. 1H-NMR, FT-IR, isothermal microcalorimetry, and NOESY NMR.12 The first two techniques were used by us to examine the interaction between bifunctional polymeric guest 2-(1550) and the dendritic host. In order to do this, two solutions in deuterated chloroform were prepared, the first containing the bifunctional guest and the second containing the dendritic host. 1H-NMR and FT-IR spectra were taken of these solutions. Subsequently, the two solutions were mixed in a 16:1 molar ratio, making sure that a
129
Chapter 7
stoichiometric ratio of both complexing groups was achieved. Again, 1H-NMR and FT-IR spectra were taken of this solution, and compared with those of the host and guest solutions. The effect of complexation on the FT-IR spectrum is shown in figure 7.2. In the carbonyl region (1800 1600 cm-1) of the spectrum of the guest, two major bands are observed. The band at 1732 cm-1 corresponds to the carbonyl of the acid functionality, and the band at 1664 cm-1 to the carbonyl of the urea group. According to this wavenumber, the urea group is non-hydrogen bonded. The host shows one band in this region at 1637 cm-1, indicating that in the periphery of the dendrimer the urea groups are hydrogen bonded.15 In the complex, the band of the carbonyl of the acid functionality has disappeared, which proves the deprotonation of this group. One band is visible at 1630 cm-1, indicating that hydrogen bonding between the urea groups has become stronger upon complexation. No shift of this band was noticed upon diluting the solution, demonstrating the robustness of the complex.
Guest 1732 1664
Host
%T
1637 Complex
1630
1800
1750
1700
1650
1600 -1 (cm )
1550
1500
1450
1400
Figure 7.2: FT-IR spectra of solutions of bifunctional guest 2-(1550), host, and their complex in chloroform, concentration is 60 g/l.
1
H-NMR spectra of bifunctional guest, host, and their complex are depicted in figure
7.3. Upon addition of the guest to the host, the peak at 2.4 ppm, corresponding to the methylene groups adjacent to tertiary amines of the host, shifts slightly downfield and broadens considerably. This is caused by protonation of these tertiary amines by the carboxylic acid functionalities of the guest. Furthermore, the peaks between 5.4 and 6.2 ppm, corresponding to the urea groups of both host and guest, shift slightly downfield and broaden, as well. This is an indication that hydrogen bonding between the dendritic host and the 130
encapsulated guest occurs. The broadening is a result of the rigidification of the periphery of the dendrimer and of an increase in (apparent) molecular weight upon complexation.
Guest
Host
Complex
4 (ppm)
Figure 7.3: 1H-NMR spectra of solutions of bifunctional guest 2-(1550), host, and their complex in CDCl3. The trends that were observed in the 1H-NMR and FT-IR spectra upon complexation of the host and guests, are typical for this supramolecular binding motif.12 Evaporation of the solution to complete dryness, gave a transparent film with a completely different appearance than the starting host and guest, which are an amorphous powder and a waxy solid, respectively. The film possesses appealing mechanical properties, such as elasticity, although it is slightly sticky.
7.4 Viscometry of polymeric guest dendritic host complexes Upon mixing chloroform solutions of the dendritic host and the bifunctional guests, a pronounced increase of the viscosity was observed. This is an indication of the formation of a network, or at least of larger aggregates. To examine this behaviour in more detail, a viscometric study of these solutions was performed, using an Ubbelohde viscometer.
131
Chapter 7
7.4.1 Polydisperse, bifunctional guests The specific viscosity, sp, of a solution is defined as: sp = 1 0
(eq. 7.1)
in which and 0 are the absolute viscosity of solution and pure solvent, respectively.16 For a solution of the dendritic host and bifunctional guest 2-(1550), the logarithm of the specific viscosity is plotted versus the logarithm of the concentration in figure 7.4. The double logarithmic scale of this plot enables us to study the association behaviour of this supramolecular system over a wide concentration range. From this plot, it is evident that the viscosity is strongly concentration dependent. At low concentrations, the viscosity is low and the slope of the curve is approximately 1, corresponding to the dilute region of polymer solutions in which no overlap of polymer coils occur.17 Thus, in this region isolated associates are present that do not interact with one another. Both ureido-acetic acid groups of one bifunctional guest are complexed to the same dendritic host (intramolecular association, scheme 7.2 left). This kind of associates is referred to as flowers-like structures. These dilute solutions are completely transparent. Upon increasing the concentration, the curve becomes steeper and the slope becomes larger than one. This is an indication that the supramolecular associates start to interact, presumably by association of both end groups of one bifunctional guest to different dendritic hosts (intermolecular complexation), causing larger aggregates to be formed (scheme 7.2 middle). In this concentration region, the solution is not transparent anymore, but slightly hazy (more details will follow). Increasing the concentration further leads to very viscous solutions, and the concentration dependency becomes very strong (slope higher than 4), demonstrating that the associates strongly interact. The end groups of one guest predominantly complex with different dendritic hosts, causing the solution to be reversibly crosslinked, and a transient network is formed (scheme 7.2 right). These viscous solutions are completely transparent, again. The intersection of the tangents at low and high concentrations is an indication of the concentration at which the associates begin to interact with one another. This concentration is analogous to the critical overlap concentration, c*, of solutions of linear polymers at which the polymer chains become entangled. For this system, the critical overlap concentration is 22 g/l, i.e. the concentration of the dendritic host equals 5.2*10-4 mol/l.
132
4.0
3.0
2.0 log sp
Slope>4
1.0
0.0
-1.0
Slope=1
-2.0 0.0 0.5
1.0 1.5 log c (g/l)
c*
2.0
2.5
Figure 7.4: Specific viscosity versus concentration of a solution of dendritic host and bifunctional guest 2-(1550) in chloroform.
The bifunctional guest with the higher molecular weight, 2-(4500), was complexed with the dendritic host, as well. The concentration dependency of the viscosity of a solution of this complex is shown in figure 7.5. Compared to the complex with 2-(1550), the curve is very similar. The critical overlap concentration for this system is 23 g/l, almost identical to that of the previous system. However, expressed in the molar concentration of dendritic host, the critical overlap concentration is considerably lower, namely 2.5*10-4 mol/l. This is in agreement with our expectations, since the higher molecular weight guest is longer, and as a result, it can more easily bridge two dendritic hosts that are farther apart, compared to the shorter guest, Hence, network-formation is favoured at lower molar concentration. The viscosity behaviour of our system is similar to that of HEUR triblock copolymers in water.4,7 However, since in those systems the molecular weights of the polymeric spacer between the hydrophobes (Mn=10000 40000 g/mol) are much higher than the molecular weights of our bifunctional guests, the critical overlap concentration in the HEUR systems is much lower.
133
Chapter 7
4.0
3.0
2.0 1-(1650) 2-(1550) 2-(4500) log sp
1.0
0.0
-1.0
-2.0 0.0 0.5 1.0 1.5 log c (g/l) 2.0 2.5
Figure 7.5: Viscosity plot of a solution of dendritic host and polymeric guests.
7.4.2 Polydisperse, monofunctional guest
At this point, the question arises whether the steep increase of the viscosity is truly caused by the formation of a reversible network, or just by the fact that a concentrated solution of macromolecules is obtained. In order to answer this question, monofunctional guest 1-(1650) was complexed with the dendritic host, and the concentration dependency of the viscosity of this solution was studied (figure 7.5). Comparing the curves corresponding to the bifunctional guests and the monofunctional guest, clearly shows the different behaviour of these complexes. The slope of the latter curve is also approximately 1 at low concentration, however, in contrast to the curve of the bifunctional guests, it hardly increases at higher concentration. Apparently, the associates of the complex with the monofunctional guest hardly interact with one another. Due to the absence of a second guest group, the complex is unable of forming a network. In our opinion, this complex can be described as hairy spheres, and can be compared with star polymers of which the arms are too short to entangle.18 It clearly demonstrates the requirement of multiple ureido-acetic acid groups per guest, for the formation of a network.
134
7.4.3 Monodisperse, bifunctional guest
The bifunctional, polymeric guests 2-(1550) and 2-(4500) are polydisperse in nature (table 7.1). The curves in figure 7.5 indicate a gradual transition from flower-like structures to a transient network, upon increasing the concentration. To study the influence of the polydispersity on the smoothness of this transition, we prepared a monodisperse, bifunctional guest, 2-(1270, MD). This guest is based on the monodisperse pTHF-oligomer that was described in chapter 6. The number-averaged molecular weight of this guest is 1270 g/mol, and its polydispersity is 1.003.
O HO O N H N H O O 13 2-(1270, MD) H N O H N O OH
To obtain a 100 g/l solution of the complex of the dendritic host and the monodisperse, bifunctional guest, the latter was dissolved in chloroform and added to a solution of the dendritic host. Instantaneously, a phase separated emulsion was formed that settled very slowly. This behaviour is very different from that of the complex containing the polydisperse, bifunctional guests, which did not show phase separation. After a few days, two distinct layers were obtained. Both layers were completely transparent and had almost equal volumes. These two layers were isolated from one another, and analyzed. The upper layer was highly viscous, and the concentration of the complex was determined at 170 g/l. The lower layer was not viscous at all, and a concentration of 30 g/l was determined. 1H-NMR and FTIR analysis of the layers showed that both layers contained the same ratio of host and guest. The viscosity of the upper, concentrated layer was too high to be measured by capillary viscometry, however, it was evident that this viscosity was much higher than the viscosity of a 100 g/l solution of the complex of dendritic host and bifunctional guest 2(1550). This upper layer is believed to contain a network structure. Addition of more solvent to this layer does not result in dilution of the concentrated solution, since it does not mix with the pure solvent. This might indicate the presence of higher-ordered species, that are not in a thermodynamic equilibrium. For the lower layer, the viscosity of a dilution series was measured (figure 7.6). The slope of this curve is close to 1, indicating the absence of entanglements or other interactions between the associates in solution. This is believed to be caused by the exclusive formation of flower-like structures in the lower layer, no larger aggregates are formed at these lower concentrations.
135
Chapter 7
4.0
3.0
2.0 log sp 2-(1550) 2-(1270, MD)
1.0
0.0
-1.0
-2.0 0.0 0.5 1.0 1.5 log c (g/l) 2.0 2.5
Figure 7.6: Viscosity plot of a solution of dendritic host and bifunctional guests.
This monodisperse, bifunctional guest demonstrates the remarkable differences that are achieved by employing well-defined building blocks, which gives rise to sharper phase transitions.
7.4.4 Temperature-dependence
The viscosity of a solution of the dendritic host and bifunctional guest 2-(1550) is highly temperature dependent, as is shown in figure 7.7. Upon increasing the temperature by 35C, the viscosity drops by a factor of 16. Representing this data in an Arrhenius plot (figure 7.7 inset), yields a straight line, and according to the Arrhenius-Frenkel-Eyring equation16: E (eq. 7.2) = A * exp( ) RT the activation energy for viscous flow, E = 64 kJ/mol. This value is significantly higher than values measured for entangled solutions of linear polymers (20 40 kJ/mol)19, of supramolecular polymers (17 kJ/mol),20 and of reversible networks based on telechelic ionomers (25 kJ/mol)5. It is comparable with values found for HEUR triblock copolymers (40 62 kJ/mol)7b,21; fluorinated HEURs show even higher activation energies (100 kJ/mol).21 The high activation energy suggests that another effect is superimposed on the normal temperature dependence of viscosity, which is governed by segmental motions. First of all, the association and dissociation of the complexing units are temperature dependent, and 136
will result in a decrease of the viscosity. Secondly, upon increasing the temperature, the conformation of the polymeric chain might change due to entropic reasons, causing the endto-end distance of the bifunctional guest to decrease.22 As a result, a transition from the reversible network to flower-like structures will occur, accompanied by an additional decrease of the viscosity.
1000
7 6.5
800
ln ( r)
6 5.5 5 4.5 4 3.5 3.00 3.20 1/T (10-3 K-1) 3.40
600 sp 400 200 0 10 20 30
40 T (C)
50
60
Figure 7.7: Temperature dependency of the viscosity of a solution of dendritic host and bifunctional guest 2-(1550) in chloroform, c=90 g/l; inset: Arrhenius plot.
7.5 Dynamic light scattering
Upon diluting a concentrated solution of the complex of dendritic host and bifunctional guest 2-(1550), a noteworthy effect was observed (see section 7.4.1). The concentrated, viscous solution was completely transparent. The reversible network extends throughout the entire solution, and no density differences are present; hence, the solution appears clear. Dilution of this solution turned it slightly hazy. In this semi-diluted region, we presume that large associates are formed that scatter light, causing the solution to appear slightly turbid. Upon diluting further, the solution became transparent again, since the associates become too small to scatter light. Dynamic light scattering (DLS)23 has been used by others to study the aggregation behaviour of ABA-triblock copolymers.24 We utilized this technique to study the size of the associates as a function of the concentration. Measurements were performed with an argon ion laser, at a wavelength of =514.5 nm and a scattering angle of =90. By DLS, the 137
Chapter 7
hydrodynamic radius, Rh, of spherical particles in dilute solutions can be determined employing the StokesEinstein equation: kBT (eq. 7.3) Rh = 6 D In which kB is Boltzmanns constant, T is the absolute temperature, is the solvent viscosity, and D is the diffusion coefficient of the particle in dilute solution, and is calculated from the relaxation rate, : (eq. 7.4) D= 2 q
4 n (eq. 7.5) sin( ) 2 In this equation, n is the refractive index of the solution, is the wavelength of the incident light, and is the scattering angle. The relaxation rate, , is the inverse of the relaxation time, r, which is obtained by fitting the normalized autocorrelation function G1(). In case of one single relaxation time, the autocorrelation function is described by:
q= G1( ) = e / r
(eq. 7.6) When the system is characterized by a distribution of relaxation times, the autocorrelation function is described by a linear combination of the individual autocorrelation functions. In summary, slow relaxation of the autocorrelation function (long relaxation times) corresponds to particles having large hydrodynamic radii. Figure 7.8 shows the normalized autocorrelation functions of solutions of the dendritic host and bifunctional guest 2-(1550) at different concentrations in chloroform. The concentration increases from the left (c = 3.2 g/l) to the right curve (c = 100 g/l). As a reference, the autocorrelation function of a 50 g/l solution of the uncomplexed dendritic host is depicted as well. Figure 7.8 clearly shows a strong dependency of the relaxation time of the autocorrelation function, r, on the concentration. For the dendritic host, all parameters necessary to determine its hydrodynamic radius using eq. 7.3 7.6 are known, and this results in Rh = 2.4 0.1 nm. This is in good agreement with experimental data previously reported on other functionalized fifth generation poly(propylene imine) dendrimers.25 Since not all parameters are known for this hostguest complex in solution, and the StokesEinstein equation is only valid for dilute solutions (below the critical overlap concentration), no hydrodynamic radii were determined for these complexes in solution. Therefore, we restrict ourselves to a discussion of the relaxation times to describe the trends observed in this system. The autocorrelation functions were fitted according to a first order exponential decay, which yields a single relaxation time. The quality of these fits was satisfactory, and the results are shown in figure 7.9. 138
With q as the scattering vector:
1
c=3.0 g/l
0.9 0.8 0.7 0.6 G1() 0.5 0.4 0.3 0.2 0.1 0 1 10 100
c=100 g/l
dendritic host + bifunctional guest r dendritic host
1000 (s)
10000
100000
1000000
Figure 7.8: Autocorrelation functions of solutions of the dendritic host and bifunctional guest 2-(1550) in chloroform (solid line; from left to right: c = 3.0, 4.8, 7.9, 12.6, 20, 35, 55, 73,
100 g/l), and of the uncomplexed dendritic host (dotted line).

10000
1000
r (s)
100
1-(1650) 2-(1550)
10
1 0.0 0.5 1.0 1.5 log c (g/l) 2.0 2.5
Figure 7.9: Relaxation times, obtained by fitting the autocorrelation functions, as a function
of the concentration. 139
Chapter 7
In case of the complex containing the bifunctional guest 2-(1550), the relaxation times seem to be constant at low concentration, indicating that the hydrodynamic radius of the particles does not increase. This supports our model which states that in dilute solution only flowers are formed. The relaxation time of these flowers is slightly higher than that of the uncomplexed dendritic host, r = 14 s versus 7.9 s, respectively, indicating that the flowers are approximately two times larger than the uncomplexed host. Increasing the concentration of the solution, results in a drastic increase of the relaxation time, demonstrating an increase in size of the particles in solution. This is in agreement with our proposed model (scheme 7.2) which predicts the occurrence of larger associates at higher concentrations, due to complexation of one bifunctional guest with two dendritic hosts. The onset of the strong increase of relaxation times is between 10 and 20 g/l, which is close to the critical overlap concentration that was determined by viscometry (23 g/l). DLS measurements were also performed on the complex of the dendritic host and the monofunctional, polymeric guest 1-(1650) (figure 7.9). For this system, the relaxation times are almost independent of the concentration, indicating the occurrence of particles of constant size over the entire concentration region. Based on the relaxation time, their size is only slightly larger than that of the uncomplexed dendritic host. This confirms our previous assumption, in which we describe the structures formed in this system as hairy spheres that are unable to form larger associates.
7.6 Conclusions and outlook
In this chapter, we have demonstrated the design of a novel type of reversible networks by employing well-defined interactions between a dendritic host containing 32 binding sites, and telechelic polymers bearing two guest groups. Spontaneous complexation takes place upon mixing bifunctional guests with the dendritic host in solution. The viscosity of a solution of this complex was highly dependent on the concentration. At low concentrations, the curve of the viscosity plot suggests the presence of associates that do not interact with one another. We describe these associates as flower-like structures in which both end groups of one telechelic guest are complexed to the same dendritic host. Increasing the concentration, results in viscous solutions of which the viscosity is highly concentration dependent. A gradual transition from flower-like structures towards larger associates, and eventually a transient network occurs. Both end groups of one telechelic guest are now complexed to two different dendritic hosts by forming a bridge between them, and hence, giving rise to a network structure.
140
The validity of this model was verified by employing also a monofunctional, polymeric guest. A solution of this guest with the dendritic host displays a much lower viscosity, and the concentration-dependency of the viscosity denotes the absence of interactions between the associates, even at high concentration. Because the monofunctional guest is unable to form a network by bridging two different hosts, these associates are referred to as hairy spheres. Dynamic light scattering measurements confirm the results obtained by viscometry. For the complex with the bifunctional guest, small particles are present at low concentration. Upon increasing the concentration, the relaxation time of the autocorrelation function increases dramatically, indicating the formation of much larger particles, as is predicted by our model. The size of the associates comprising the dendritic host and the monofunctional guest does not depend on the concentration, and is comparable to the size of the uncomplexed dendritic host. A bifunctional guest, 2-(1270, MD), was prepared with a monodisperse molecular weight distribution, in order to study the influence of the polydispersity on the transition from flowers to a network. Its molecular weight and structure are comparable to the polydisperse bifunctional guest 2-(1550), but as it was complexed with the dendritic host, it behaved very differently. A solution of this complex phase separated into two layers: a dilute, low-viscosity layer and a concentrated, highly viscous layer. This phase separation shows strong resemblance with the sol-gel transition that is often observed in polymer solutions.26 We propose that the former layer contains flowers exclusively, while the latter layer contains the network structure. Due to the monodisperse nature of this bifunctional guest, a sharp transition between the flower-like structures and the network occurs. Until so far, only a minor part of the characteristics of these reversible networks has been studied. In our opinion, many more aspects can be varied that influence the properties of this system. For example, by changing the generation of the dendrimer (the number of host units per dendrimer) or by incorporating monofunctional guests, the functionality of the reversible crosslinks can be tuned. Furthermore, by varying the molecular weight of the bifunctional guest, the crosslink-density and critical overlap concentration are changed. The polarity of the polymer backbone of the guest, and of the solvent have a major effect on the solution properties, and it is very appealing to generate these reversible networks in aqueous media. Some first approaches to this goal, by replacing the main chain of the guest by a water soluble, poly(ethylene oxide) backbone, are promising. However, the complexation in water seems less straightforward than in organic solvents. It is highly interesting to study the rheological properties of the complex solutions at different shear rates, since comparable systems (the ABA-triblock copolymers described in the introduction) are known to show both
141
Chapter 7
shear-thinning and shear-thickening.7 Finally, the bulk properties of these reversible networks have to be explored in further detail. This system is very suitable to be used in a modular approach, as is described in section 5.1. By complexing the dendritic host with a combination of bifunctional guests and guests bearing additional functionalities, a network will be formed that has elastic properties as well as functionality. The number of possible applications this could yield, is nearly unlimited.

General methods and instrumentation. General methods concerning purification of solvents and spectroscopic techniques can be found in the experimental section of previous chapters. Solution viscosities were measured with a Schott-Gerte ubbelohde micro-viscometer with a suspended level bulb of type 538/30: K=1, 538/20: K=0.1 and 538/10: K=0.01. The sample was thermostated with a bath of type CT1450, and elution times were measured with a Schott AVS 350. Dynamic light scattering (DLS): The scattering cells (5 ml cylindrical ampules) were immersed in a largediameter thermostated bath of index-matching liquid (water) at 20C. The DLS measurements were performed using a Spectra-Physics model 2020 Ar ion laser, operating at 514.5 nm, and variable output power. Typical laser power was 50 mW. The scattering angle was 90, and the detector optics employed a Malvern Autosizer 4700 with 256 linearly spaced channels. Solvents were filtered through a 0.45 m membrane filter prior to measurements. Materials. The availability of some chemicals was discussed in the experimental section of previous chapters. The dendritic host, DAB-dendr-(NHCONH-Ad)64, was kindly provided by Bas de Waal; methyl trifluoromethanesulfonate and ethyl isocyanatoacetate were purchased from Aldrich. -(4-Aminobutyl)--methyl-poly(tetrahydrofuran) Freshly distilled tetrahydrofuran (10.0 ml, 0.123 mol) was inserted into a dried 50 ml flask via a syringe under exclusion of air, and 2,6-di-tert-butylpyridine (0.75 ml, 3.35 mmol) was added to this. The solution was cooled to 0C and kept under an argon atmosphere. Methyl trifluoromethanesulfonate (1.00 g, 6.09 mmol) was dissolved in dichloromethane (10 ml), and added to the previous solution. The reaction mixture was stirred for 50 min at 0C under an argon atmosphere. The reaction was quenched by addition of hexamethylene tetraamine (2.14 g, 15.3 mmol) dissolved in chloroform (100 ml). The turbid reaction mixture was allowed to warm to room temperature. Subsequently, it was evaporated to dryness under reduced pressure, and the residue was redissolved in methanol (60 ml) and toluene (40 ml), and sulfuric acid (3 ml) was added. The reaction mixture was heated to 100C for 3 h, and after cooling down, it was neutralized by addition of concentrated sodium hydroxide solution (approx. 3 ml 10 M). The salts were filtered off, and the filtrate was evaporated to dryness under reduced pressure, redissolved in THF (10 ml), and precipitated in cold, diluted sodium hydroxide solution (100 ml 0.1 M). The white precipitate was filtered off, washed, and redissolved in diethyl ether (50 ml). The solution was dried with sodium sulfate and then evaporated to dryness under reduced pressure while cooling the flask on an ice bath, allowing a white powder to crystallize (2.28 g). 1H-NMR (CDCl3, 400 MHz): 3.41 (m, 62H, CH2CH2O), 3.33 (s, 3H, CH3), 2.71 (t, 2H, CH2NH2, J=5.5 Hz), 1.62 (m, 62H, OCH2CH2), 1.50 (m, 2H, CH2CH2NH2). 13C-NMR (CDCl3, 100 MHz): 72.6 (CH3OCH2), 70.5 (CH2CH2O), 58.5 (CH3O), 42.1 (CH2NH2), 30.6 (CH2CH2NH2), 26.5 (OCH2CH2). Mn(1H-NMR)=1150 g/mol.
142
-[4-(Carboxymethylureido)butyl]--methyl-poly(tetrahydrofuran), 1-(1650) -Amino--methoxy-poly(tetrahydrofuran) (0.50 g, 0.43 mmol) was dissolved in chloroform (10 ml) and ethyl isocyanatoacetate (61.8 mg, 0.48 mmol) was added. The reaction mixture was stirred at room temperature for 30 min, and subsequently evaporated to dryness. The product was redissolved in THF (15 ml) and sodium hydroxide solution (15 ml 1 M) was added. The reaction mixture was stirred overnight, after which it was slightly acidified by addition of concentrated hydrochloric acid solution (approx. 1.5 ml), and concentrated in vacuo. The aqueous layer was extracted with dichloromethane (three times 20 ml), and the organic layers were combined, dried with sodium sulfate, and evaporated to dryness. The product was obtained as a pale, waxy solid (0.47 g, 66%). 1H-NMR (CDCl3): 5.32 (br. s, 1H, NH), 5.18 (br. s, 1H, NH), 3.95 (s, 2H, CH2COOH), 3.41 (m, 84H, CH2CH2O), 3.33 (s, 3H, CH3), 3.22 (m, 2H, CH2CH2NH), 1.62 (m, 84H, OCH2CH2), 1.43 (m, 2H, CH2CH2NH). FT-IR (ATR): 3331 (NH stretching), 2938, 2853, 1751 (COOH), 1647 (C=O urea), 1563, 1447, 1369, 1105 cm-1. Mn(1H-NMR)=1650 g/mol. Bis[3-(carboxymethylureido)propyl]-poly(tetrahydrofuran), 2-(x) The synthesis of 2-(1550) is given as a typical example. Bis(3-aminopropyl)-poly(tetrahydrofuran), Mn=1100 g/mol, (2.00 g, 1.82 mmol) was dissolved in chloroform (40 ml) and ethyl isocyanatoacetate (0.47 g, 3.64 mmol) was added. The reaction mixture was stirred at room temperature for 30 min, and subsequently evaporated to dryness. The product was redissolved in THF (20 ml) and sodium hydroxide solution (20 ml 1 M) was added. The reaction mixture was stirred for 2 days, after which it was slightly acidified by addition of concentrated hydrochloric acid solution (approx. 2 ml), and concentrated in vacuo. The aqueous layer was extracted with dichloromethane (three times 25 ml), and the organic layers were combined, dried with sodium sulfate, and evaporated to dryness. The product was obtained as a yellow, waxy solid (2.10 g, 90%). 1H-NMR (CDCl3): 5.6 (br. s, 4H, NH), 3.94 (s, 4H, CH2COOH), 3.51 (t, 4H, OCH2CH2CH2NH), 3.41 (m, 68H, CH2CH2O), 3.28 (t, 4H, CH2CH2NH), 1.76 (m, 4H, OCH2CH2CH2NH), 1.62 (m, 84H, OCH2CH2). FT-IR (ATR): 3359 (NH stretching), 2939, 2855, 1733 (COOH), 1638 (C=O urea), 1568, 1368, 1103 cm-1. Mn(1H-NMR)=1550 g/mol. Preparation of the reversible network The dendritic host, DAB-dendr-(NHCONH-Ad)64, (0.214 g, 1.15*10-5 mol) was dissolved in chloroform (2 ml), and added to a solution of 2-(1550) (0.286 g, 1.85*10-4 mol) in chloroform (2 ml). The volume of the solution was filled to 5.00 ml to obtain a 100 g/l solution of the complex. The viscous solution was thoroughly mixed by mechanical stirring and/or sonication. FT-IR: 3356 (NH stretching), 1630 (C=O urea), 1522 cm-1.

1. a) Section 1.1 of this thesis; b) Legge, N. R., Holden, G. and Schroeder, H. E. Thermoplastic elastomers: A comprehensive review; Carl Hansser Verslag: New York, 1987. 2. Holden, G., Bishop, E. T. and Legge, N. R. J. Pol. Sci. C 1969, 26, 36. 3. Quintana, J. R., Janez, M. D. and Katime, I. Polymer 1998, 39, 2111. 4. a) Kaczmarski, J. P. and Glass, J. E. Macromolecules 1993, 26, 5149; b) Glass, J. E. J. Coat. Tech. 2001, 73, 79; c) Xu, B., Yekta, A., Li, L., Masoumi, Z. and Winnik, M. A. Colloid. Surf. 1996, 112, 239. 5. Chassenieux, C., Nicolai, T., Tassin, J.-F., Durand, D., Gohy, J.-F. and Jerome, R. Macromol. Rapid Commun. 2001, 22, 1216. 6. a) Egland-Jongewaard, S. K. and Glass, J. E. Polym. Mater. Sci. Eng. 1984, 50, 485; b) Thibeault, J. C., Sperry, P. R. and Schaller, E. J. Polym. Mater. Sci. Eng. 1984, 51, 353; c) Wang, Y. and Winnik, M. A. Langmuir 1990, 6, 1437; d) Yekta, A., Duhamel, J., Brochard, P., Adiwidjaja, H. and Winnik, M. A. Macromolecules 1993, 26, 1829; e) Yekta, A., Duhamel, J., Adiwidjaja, H., Brochard, P. and Winnik, M. A. Langmuir 1993, 9, 881. 7. a) Tam, K. C., Jenkins, R. D., Winnik, M. A. and Bassett, D. R. Macromolecules 1998, 31, 4149; b) Ma, S. X. and Cooper, S. L. Macromolecules 2002, 35, 2024. 8. a) Menchen, S., Johnson, B., Winnik, M. A. and Xu, B. Electrophoresis 1996, 17, 1451; b) Menchen, S., Johnson, B., Winnik, M. A. and Xu, B. Chem. Mater. 1996, 8, 2205.
143
Chapter 7
9. a) Tae, G., Kornfield, J. A., Hubbell, J. A., Johannsmann, D. and Hogen-Esch, T. E. Macromolecules 2001, 34, 6409; b) Tae, G., Kornfield, J. A., Hubbell, J. A. and Lal, J. Macromolecules 2002, 35, 4448. 10. Folmer, B. J. B., Brunsveld, L., Meijer, E. W. and Sijbesma, R. P. Chem. Rev. 2001, 101, 4071. 11. a) Yamauchi, K., Lizotte, J. R., Hercules, D. M., Vergne, M. J. and Long, T. E. JACS 2002, 124, 8599; b) Lohmeijer, B. G. G. and Schubert, U. S. Angew. Chem. Int. Ed. 2002, 41, 3825; c) Hirschberg, J. H. K. K., Sijbesma, R. P., Ramzi, A. and Meijer, E. W. Macromolecules , submitted. 12. a) Baars, M. W. P. L., Karlsson, A. J., Sorokin, V., De Waal, B. F. W. and Meijer, E. W. Angew. Chem. Int. Ed. 2000, 39, 4262; b) Boas, U., Karlsson, A. J., de Waal, B. F. M. and Meijer, E. W. J. Org. Chem. 2001, 66, 2136; c) Boas, U., Sontjens, S. H. M., Jensen, K. J., Christensen, J. B. and Meijer, E. W. ChemBioChem 2002, 3, 433. 13. Flory, P. J. Principles of Polymer Chemistry; Cornell University Press: Ithaca: New York, 1953. 14. a) Dubreuil, M. F. and Goethals, E. J. Macromol. Chem. Phys. 1997, 198, 3077; b) Dubreuil, M. F., Farcy, N. G. and Goethals, E. J. Macromol. Rapid Commun. 1999, 20, 383; c) Goethals, E. J., Caeter, P. V., Geeraert, J. M. and Prez, F. E. D. Ang. Makromol. Chem. 1994, 223, 1; d) Oike, H., Yoshioka, Y., Kobayashi, S., Nakashima, M., Tezuka, Y. and Goethals, E. J. Macromol. Rapid Commun. 2000, 21, 1185; e) Vanrenterghem, T., Dubreuil, M. F., Goethals, E. J. and Loontjens, T. J. Polymer International 1999, 48, 343. 15. Baars, M. W. P. L. and Meijer, E. W. Polym. Mater. Sci. Eng. 1997, 77, 149. 16. Hiemenz, P. C. Polymer Chemistry - The basic concepts; Marcel Dekker, Inc.: New York, 1984. 17. a) Elias, H.-G. An Introduction to Polymer Science; VCH: Weinheim, 1997; b) Milas, M., Rinaudo, M., Knipper, M. and Schuppiser, J. L. Macromolecules 1990, 23, 2506. 18. Fetters, L. J., Kiss, A. D., Pearson, D. S., Quack, G. F. and Vitus, F. J. Macromolecules 1993, 26, 647. 19. a) Sinha, V. K., Patel, K. H. and Trivedi, H. C. Angew. Makromol. Chem. 1988, 157, 31; b) Patel, B. K., Sinha, V. K., Makhija, K. K., Ray, A. and Trivedi, H. C. J. Macromol. Sci., Phys. 1990, B29, 397; c) Pezzin, G. and Gligo, N. J. Appl. Polymer Sci. 1966, 10, 1; d) Okada, R. and Tanzawa, H. J. Pol. Sci. A 1965, 3, 4294. 20. Folmer, B. J. B. Ph.D. Thesis, Eindhoven University of Technology (Eindhoven), 2000. 21. Cathebras, N., Collet, A., Viguier, M. and Berret, J.-F. Macromolecules 1998, 31, 1305 22. Painter, P. C. and Coleman, M. M. Fundamentals of polymer science; 2 ed.; Technomic: Lancaster, 1997. 23. a) Berne, B. J. and Pecora, R. Dynamic Light Scattering; Wiley-Interscience: Chichester, 1976; b) Pecora, R. Dynamic Light Scattering - Applications of Photon Correlation Spectroscopy; Plenum Press: New York, 1985; c) Schmitz, K. S. An introduction to Dynamic Light Scattering by Macromolecules; Academic Press: San Diego, 1990. 24. a) Dai, S., Tam, K. C., Jenkins, R. D. and Bassett, D. R. Macromolecules 2000, 33, 7021; b) Gohy, J.-F., Antoun, S. and Jerome, R. Macromolecules 2001, 34, 7435; c) Suwa, M., Hashidzume, A., Morishima, Y., Nakato, T. and Tomida, M. Macromolecules 2000, 33; d) Xu, R., Winnik, M. A., Hallett, F. R., Riess, G. and Croucher, M. D. Macromolecules 1991, 24, 87; e) Schillen, K., Brown, W. and Konak, C. Macromolecules 1993, 26, 3611. 25. Baars, M. W. P. L. Ph.D. Thesis, Eindhoven University of Technology (Eindhoven), 2000. 26. Martin, J. E. and Adolf, D. Annu. Rev. Phys. Chem. 1991, 42, 311.
144
Summary
Since their discovery in the 1940s, thermoplastic elastomers (TPEs) have drawn extensive attention from polymer scientists, because of their attractive properties. They possess elastomeric, rubber-like properties at low temperatures, while they exhibit flow at elevated temperatures, like thermoplastics do. One of the most appealing TPEs is spandex (Lycra), a highly elastomeric fiber, that is used extensively in textile fabrics. Due to the synthetic procedure to prepare TPEs, their microstructure is difficult to control, resulting in complex morphologies. A beautiful example of a TPE that is found in Nature, is spider dragline silk. The exceptional toughness of this biopolymer is a direct result of the precise control the spider has over the molecular structure of the protein, the secondary interactions within the material, and the spinning process of the fiber. From this example, we can learn that exact control over the molecular structure (uniform and monodisperse building blocks) and supramolecular interactions (hydrogen bonding, electrostatic interactions) can give rise to novel materials possessing unique properties, and is crucial to get a better understanding of the structure properties relationship. In this thesis, we applied highly-selective chemistry and precisely-controlled supramolecular interactions to construct well-defined thermoplastic elastomers. This concept was employed on two different systems: on block copoly(ether urea)s and on reversible networks based on dendrimer telechelic associations. The synthesis and characterization of these systems is described, together with a detailed study of their morphology and mechanical properties. A key role to reach this aim is the use of a recently disclosed synthetic procedure for the preparation of isocyanates, by reacting amines with di-tert-butyl tricarbonate. By employing this chemistry, it was possible to synthesize monomers that were previously inaccessible. In chapter 2, the preparation of -isocyanato--alcohols is described. Subsequent polymerization of these monomer gives the [n]-polyurethanes, a class of polymers that was still unknown until now. Hyperbranched polyurethanes were also prepared via this procedure. Block copoly(ether urea)s with uniform microstructures were synthesized from amineterminated prepolymers (chapter 3). Using protective group chemistry, the number of urea groups in the hard blocks of these polymers was exactly defined. Hydrogen bonding between the urea groups leads to the formation of long stacks of urea groups embedded in the amorphous polyether phase. These stacks serve as reversible crosslinks, giving the material elastomeric properties. Chapter 4 describes the properties of these polymers. The morphology was studied in detail by a combination of techniques, such as FT-IR, DSC, AFM and several 145
Summary
X-ray scattering techniques. These materials were completely transparent, and showed very interesting tensile properties; they could be elongated to 10 20 times their initial length. These well-defined polymers proved to be superior compared to a less-defined analogue possessing a polydisperse hard block. The uniformity of the hard block of the block copoly(ether urea)s enabled us to anchor molecules containing the complementary bisureido-butylene unit to these polymers, since they are tightly bound by supramolecular interactions. This allowed us to utilize this recognition site for the reversible functionalization of the polymeric material, and hence, this approach opens the way towards a modular methodology of polymer modification, as is illustrated in chapter 5. In chapter 6, the concept of well-defined block copolymers is taken one step further by the introduction of a monodisperse soft block. For this purpose, a monodisperse prepolymer was synthesized by a stepwise procedure, resulting in a pTHF 13-mer (MW=954 g/mol) of very low polydispersity (PD=1.004). This oligomer was used for the preparation of a block copoly(ether urea) possessing a uniform hard and a monodisperse soft block. Its properties were compared to those of the analogous polymer with the polydisperse soft block. Whereas the latter polymer is highly soluble in chloroform, the well-defined polymer forms a gel, indicating the presence of more order in this structure. The tensile behaviour of the welldefined polymer showed a more pronounced strain hardening effect. A second system, which was studied in chapter 7, is based on the complexation of the multiple binding sites of a dendritic host with the end groups of a bifunctional polymeric guest. This complexation results from hydrogen bonding between urea groups and electrostatic interactions. The behaviour of this system in solution was studied by FT-IR, 1HNMR, viscometry and dynamic light scattering. The viscosity of a solution of this complex is highly dependent on the concentration and temperature. This is caused by the presence of flower-like aggregates in dilute solutions, and the formation of a transient network in concentrated solutions. A monodisperse, bifunctional, polymeric guest was prepared and complexed with the dendritic host, and this system showed very different behaviour: phase separation was observed. This again indicates the presence of more ordering and sharper phase transitions in well-defined systems. In summary, we demonstrated the possibility to prepare well-defined thermoplastic elastomers by applying highly-selective chemistry and precisely-controlled supramolecular interactions. These materials possess remarkable properties, that differ from those of their illdefined analogues. From now on, one can study these systems in more detail and make use of their added value over the less well-defined counterparts.
146
Samenvatting
Sinds hun ontdekking in de jaren veertig hebben thermoplastische elastomeren (TPEs), wegens hun aantrekkelijke eigenschappen, de aandacht getrokken van polymeerkundigen. Zij vertonen elastomere, rubberachtige eigenschappen bij lage temperaturen, terwijl ze vloeien bij hogere temperaturen, net zoals thermoplasten. En van de meest aantrekkelijke TPEs is spandex (Lycra), een erg elastische vezel die veel gebruikt wordt in de textiel industrie. De moleculaire structuur van veel TPEs is vaak slechts matig gedefinieerd. Dit wordt veroorzaakt door de synthetische procedure via welke deze gemaakt worden, resulterende in complexe morfologin. Spinnenzijde is een prachtig voorbeeld van een TPE die in de natuur voorkomt. De buitengewone taaiheid van dit biopolymeer is het resultaat van de precieze controle die de spin uitoefent op de microstructuur van het eiwit, op de secundaire interacties in het materiaal en op de condities waaronder het vezelspinnen plaatsvindt. Dit voorbeeld leert ons dat exacte controle over de moleculaire structuur (uniforme en monodisperse bouwstenen) en over de supramoleculaire interacties (waterstofbruggen, elektrostatische interacties) kunnen leiden tot nieuwe materialen met unieke eigenschappen, en van cruciaal belang is om een beter inzicht te verkrijgen in de relatie tussen structuur en eigenschappen. In dit proefschrift zijn selectieve chemie en precies-gecontroleerde supramoleculaire interacties gebruikt om goedgedefinieerde thermoplastische elastomeren te maken. Dit concept werd toegepast op twee systemen: op blok copoly(ether ureum)s en op reversibele netwerken gebaseerd op de associatie tussen dendrimeren en bifunctionele polymeren. De synthese en karakterisering van deze systemen wordt beschreven, samen met een gedetailleerde studie van hun morfologie en mechanische eigenschappen. Een vooraanstaande rol met betrekking tot dit doel is weggelegd voor een nieuwe syntheseroute voor isocyanaten, m.b.v. di-tert-butyl tricarbonaat. Hiermee werd het mogelijk om totnogtoe ontoegankelijke monomeren te synthetiseren. In hoofdstuk 2 wordt de synthese van -isocyanato--alcoholen beschreven. Vervolgens werden deze gepolymeriseerd, hetgeen resulteerde in de [n]-polyurethanen, een vooralsnog onbekende klasse van polymeren. Via deze procedure werden ook hoogvertakte polyurethanen gemaakt. Blok copoly(ether ureum)s met een uniforme microstructuur werden gesynthetiseerd uitgaande van amine-getermineerde prepolymeren (hoofdstuk 3). Het aantal ureum groepen in de harde blokken van deze polymeren werd exact bepaald door gebruik te maken van beschermende groepen. Waterstofbruggen tussen de ureumgroepen leiden tot de vorming van reversibele knooppunten en maken het materiaal elastisch. Hoofdstuk 4 beschrijft de eigenschappen van deze polymeren. Een scala aan technieken, zoals FT-IR, DSC, AFM en verscheidene Rntgen-verstrooiingstechnieken, werden gebruikt om de morfologie in detail te 147
Samenvatting
bestuderen. Deze materialen zijn volkomen transparant en vertonen interessante rekeigenschappen; ze kunnen tot 10 20 maal hun eigen lengte worden verstrekt. Deze goedgedefinieerde polymeren bevatten betere eigenschappen vergeleken met een analoog polymeer met een polydisperse hard blok. De uniformiteit van de harde blokken in de blok copoly(ether ureum)s maakte het mogelijk om moleculen met een complementaire bisureido-butyleen eenheid d.m.v. supramoleculaire interacties in deze polymeren te verankeren. Dit opent de weg naar een reversibele functionalisering en modificatie van het materiaal via een modulaire aanpak, zoals in hoofdstuk 5 wordt gellustreerd. In hoofdstuk 6, wordt het concept van de goedgedefinieerde blok copolymeren een stap verder genomen door de introductie van een monodisperse zacht blok. Voor dit doel werd via een stapsgewijze synthese een monodisperse prepolymeer gesynthetiseerd, hetgeen resulteerde in een pTHF 13-meer (MW=954 g/mol) met een erg lage polydispersiteit (PD=1.004). Uitgaande van dit oligomeer werd een blok copoly(ether ureum) gemaakt met een uniform hard en een monodispers zacht blok, en de eigenschappen hiervan zijn vergeleken met die van een analoog polymeer met een polydispers zacht blok. Terwijl dit laatste polymeer goed oplosbaar is in chloroform, vormt het goedgedefinieerde polymeer een gel, duidende op de aanwezigheid van meer orde in deze structuur. Het rekgedrag van het goedgedefinieerde polymeer vertoonde een sterkere mate van strain-hardening. Een tweede systeem, dat beschreven staat in hoofdstuk 7, is gebaseerd op de complexatie van de eindgroepen van bifunctionele, polymere gasten met een gastheer gebaseerd op een dendrimeer. Deze complexatie is het resultaat van waterstofbruggen tussen ureum groepen en een elektrostatische interactie. Het gedrag van dit systeem in oplossing werd bestudeerd met FT-IR, 1H-NMR, viscometrie en dynamische lichtverstrooiing. De viscositeit van een oplossing van dit complex is sterk afhankelijk van de concentratie en de temperatuur. Dit wordt veroorzaakt door de aanwezigheid van bloemachtige aggregaten in verdunde oplossingen, ten opzichte van de vorming van een netwerk in geconcentreerde oplossingen. Een monodisperse, bifunctionele, polymere gast werd gesynthetiseerd en gecomplexeerd met de dendrimere gastheer. Dit systeem vertoonde een geheel ander gedrag waarbij fasescheiding werd waargenomen. Dit is opnieuw een indicatie voor de aanwezigheid van meer orde en scherpere fase overgangen in het geval van de goedgedefinieerde systemen. Samenvattend: we hebben laten zien dat het mogelijk is om goedgedefinieerde thermoplastische elastomeren te maken, door gebruik te maken van selectieve chemie en precies-gecontroleerde supramoleculaire interacties. Deze materialen vertonen interessante eigenschappen die afwijken van hun minder goedgedefinieerde analoga. Momenteel kunnen deze systemen in detail worden bestudeerd en kan er gebruik worden gemaakt van hun toegevoegde waarde. 148
Dankwoord
Zonder de hulp en steun van velen had ik mijn promotie-onderzoek nooit tot een succesvol einde kunnen brengen en was de afgelopen tijd nooit zo plezierig geweest. Ik wil mijn proefschrift beindigen met hen te bedanken. Het vertrouwen en de vrijheid die mijn promotor Bert Meijer mij heeft geschonken, hebben mij gemaakt tot de wetenschapper die ik nu ben. Niets werkt zo motiverend als het enthousiasme waarmee jij wetenschappelijk onderzoek nastreeft. Ook de begeleiding van mijn copromotor Rint Sijbesma heb ik erg gewaardeerd. Jouw ideen en suggesties waren altijd erg welkom en onze discussies hebben me steeds aangezet om nog eens kritisch naar de resultaten te kijken. Jef Vekemans bedank ik voor het minutieus doorspitten van mijn proefschrift en voor de vele malen dat je ruim de tijd nam om probleempjes op het synthetische vlak op te lossen. Prof. Chris Hunter, I want to thank you for coming all the way to the Netherlands to take part in my committee. Prof. Cor Koning wil ik graag bedanken voor het corrigeren van mijn manuscript en de interesse die U in mijn onderzoek getoond heeft. Brigitte Folmer, tijdens mijn afstuderen bij jou heb je mij laten zien hoe leuk promoveren kan zijn. Jouw manier van onderzoek doen, door met 101 dingen tegelijkertijd bezig te zijn heeft me erg genspireerd. Rob Peerlings, zonder jouw pionierende werk op het gebied van de tricarbonaat-chemie was een snelle start van mijn onderzoek niet mogelijk geweest. Richard van Someren en Bas de Waal, bedankt voor het maken van de vele grammen van dit witte goud. En van de leukste taken behorende bij het promoveren is het begeleiden van studenten. Jos Paulusse, het was indrukwekkend zoals jij als research-stagiaire zelfstandig invulling gaf aan de samenwerking met DSM Research. Ik wens je veel succes met je eigen promotie. Joris Meijer, met jouw onderzoek wilden we eigenlijk twee vliegen in n klap slaan. Alhoewel dit niet helemaal lukte, heeft het genoeg aanknopingspunten opgeleverd om ermee verder te gaan. Susan Soons, jij kwam op het einde van mijn promotie bij Rolf en mij afstuderen. Het onderzoek was zeker niet eenvoudig, maar toch heb je het met veel doorzettingsvermogen volbracht. Samen met de practicanten Tim van Rens, Jos de Bock, Bart Hamers en Chris Schiepers hadden Edda Neuteboom en ik ons eigen research-groepje. Jullie begeleiden was voor ons erg leerzaam. Chris, ik ben je erg dankbaar dat jij je zomervakantie ervoor opofferde om het pTHF 13-meer te synthetiseren. Joost van Dongen, Ralf Bovee, en Xianwen Lou wil ik bedanken voor de vele analyses die ze voor me hebben uitgevoerd. Helaas wordt het belang van de analytische chemie te vaak onderschat. Speciale dank gaat uit naar Hanneke Veldhoen, Ingrid Dirkx,
149
Dankwoord
Emma Eltink, Henk Eding, Hannie van der Lee en Hans Damen, zonder wie onze vakgroep nooit zo geolied zou lopen. I want to thank Anil Kumar for his enthusiastic stimulations concerning the hyperbranched polyurethanes. Maurice Baars, Annika Karlsson, Jrn Christensen, Michael Pittelkow and Ulrik Boas are acknowledged for their help with the dendritic-clicking. Kurt Wostyn en Koen Clays van de Katholieke Universiteit Leuven ben ik zeer erkentelijk voor hun hulp bij de lichtverstrooiings-experimenten aan de dendritische netwerken. Dankzij de X-ray metingen in samenwerking met Huub Kooijman, Han Goossens en Ralf Kleppinger, en de AFM-studies door Rob Lammertink, hebben we veel meer inzicht gekregen in de morfologie van onze materialen. Otto van Asselen ben ik erkentelijk voor zijn assistentie bij de infrarood-studies. Leon Govaert, bedankt voor je hulp bij het uitvoeren van de mechanische karakteriseringen. Jij hebt me laten zijn dat naast het maken van materialen, het stukmaken ervan zeker zo leuk is. Ronald Lange, Lilian Teuwen en Alain Hilberer van DSM Research wil ik bedanken voor hun samenwerking met Jos. Ronald, jouw bovengemiddelde interesse voor mijn onderzoek, de immer voortdurende stroom van ideen en de bijbehorende discussies heb ik altijd erg op prijs gesteld. Verder bedank ik alle leden van ons supramoleculaire chemie-cluster, in het bijzonder Marcel van Genderen, voor de kritische houding tijdens de lunches en de (meestal) goede raad. Bedankt ook, alle labgenoten voor de plezierige werksfeer op lab 4, en mijn kamergenoten Henk, Cristina, Alicia, Francesca, Abdel, Rolf en Pascal voor alle gezelligheid door de jaren heen. Henk Janssen, jouw nuchtere visie op bepaalde zaken waardeer ik zeer. Ik ben onder de indruk zoals jij SyMO-Chem hebt neergezet. Naar verwachting ligt er een grote toekomst in het verschiet, maar n miljoen euro is wel een heleboel geld. Rolf, nu jij verder gaat waar mijn onderzoek is blijven liggen, wens ik je veel geluk en wijsheid bij het vervolg ervan. Natuurlijk wil ik ook alle andere (oud)-collegas bedanken die hebben bijgedragen aan een onvergetelijke tijd. Bas en Marcel, dankzij jullie kijk ik terug op een geweldige studietijd. Ik vind het fantastisch dat jullie mijn paranimfen willen zijn. Ik bedank mijn familie en vrienden, die ik helaas afgelopen tijd te kort heb gedaan. Hiervoor bied ik mijn oprechte verontschuldigingen aan. Pap, mam en Arjan, bedankt voor jullie rotsvaste vertrouwen in een goede afloop. Jullie stonden altijd voor me klaar. Dorien, tenslotte wil ik jou bedanken voor alle begrip, steun, geluk en liefde in de afgelopen jaren. Dankjewel en veel succes!
150
Curriculum Vitae
Ron Versteegen werd geboren op 30 maart 1974 te Tegelen. Na het doorlopen van een VWO-opleiding aan het Boschveld college in Venray, werd er in 1992 begonnen met de studie Scheikundige Technologie aan de Technische Universiteit Eindhoven, alwaar het propadeutisch examen werd behaald in 1993. Het afstudeerwerk met als titel Supramolecular polymeric architectures werd verricht in de vakgroep Macromoleculaire en Organische Chemie (prof.dr. E.W. Meijer). Het afsluitend examen werd in 1998 cum laude behaald. Vervolgens werd in dezelfde vakgroep als AIO begonnen aan een promotieonderzoek onder leiding van dr. R.P. Sijbesma en prof.dr. E.W. Meijer. De belangrijkste resultaten van dit onderzoek staan beschreven in dit proefschrift.
Ron Versteegen was born in Tegelen, the Netherlands, on March 30th, 1974. In 1992 he obtained his high school degree at the Boschveld college in Venray. He continued with the study of Chemical Engineering at the Eindhoven University of Technology. His major was Macromolecular and Organic Chemistry (Prof.dr. E.W. Meijer) and he graduated cum laude in 1998. Subsequently, he started as a PhD-student in the same department under the guidance of dr. R.P. Sijbesma and prof.dr. E.W. Meijer. The results from this study are described in this thesis.
151

Thermoplastic Elastome

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Thermoplastic Elastome

Caricato da

Copyright:

Formati disponibili

Well-defined Thermoplastic Elastomers

Reversible networks based on hydrogen bonding

Well-defined Thermoplastic Elastomers

Ronny Mathieu Versteegen

Copromotor: dr. R.P. Sijbesma

CIP-DATA LIBRARY TECHNISCHE UNIVERSITEIT EINDHOVEN

3.6 3.7 3.8 3.9

7.5 7.6 7.7 7.8

Summary Samenvatting Dankwoord Curriculum Vitae

Thermoplastic elastomers and well-defined polymers

Figure 1.4: Schematic representation of a TPU and its building blocks.

Thermoplastic elastomers and well-defined polymers

O Cl - HCl R N H Cl - HCl R N=C=O

Thermoplastic elastomers and well-defined polymers

Nylon 6,6 Kevlar 49 Dyneema Steel Spider silk

5 130 110 200 10

0.95 3.6 3.4 1.5 1.1

Thermoplastic elastomers and well-defined polymers

Thermoplastic elastomers and well-defined polymers

Thermoplastic elastomers and well-defined polymers

1.8 References and notes

[n]-Polyurethanes and hyperbranched polyurethanes

O HO (CH 2) N O N 2H 4 EtOH HO (CH 2 ) NH 2

[n]-Polyurethanes and hyperbranched polyurethanes

HOOC (CH 2) NH 3Cl

EtOOC (CH2) NH3 Cl

LiAlH 4 HO (CH2 ) NH 2 x THF 2

(CH 2)x x=23

Scheme 2.7: Polymerization of isocyanato alcohols to [n]-polyurethanes. 20

[n]-Polyurethanes and hyperbranched polyurethanes

0.22 0.24 0.24 0.43 0.30 0.18 0.26

[n]-Polyurethanes and hyperbranched polyurethanes

Figure 2.4: TGA of [6]-polyurethane.

[n]-Polyurethanes and hyperbranched polyurethanes

[n]-Polyurethanes and hyperbranched polyurethanes

[n]-Polyurethanes and hyperbranched polyurethanes

2.8 Experimental Section

[n]-Polyurethanes and hyperbranched polyurethanes

[n]-Polyurethanes and hyperbranched polyurethanes

[n]-Polyurethanes and hyperbranched polyurethanes

2.9 References and notes

[n]-Polyurethanes and hyperbranched polyurethanes

Synthesis and characterization of block copoly(ether urea)s

Synthesis and characterization of block copoly(ether urea)s

Synthesis and characterization of block copoly(ether urea)s

Figure 3.5: 1H-NMR spectrum of bis(3-aminopropyl)-pTHF2000 3. 45

Mn,after (g mol ) 350 1100 2500 5200 4500 2300 3600

Mw/Mn 1.8 1.7 1.5 1.5 1.4 n.d. n.d.

2000 2900 4000 2000 3500

Synthesis and characterization of block copoly(ether urea)s

MeOH / H2SO4 Toluene, reflux

Synthesis and characterization of block copoly(ether urea)s

0 0 0.02 0.04 0.06 [I]/[M] 0.08 0.1

Synthesis and characterization of block copoly(ether urea)s

7 OCN pTHF NCO + H2N R NH2 B

Synthesis and characterization of block copoly(ether urea)s

Synthesis and characterization of block copoly(ether urea)s

O H2N pTHF NH2 + H2N NH2 O pTHF N H N H 2 n + 2 O O

Scheme 3.8: Synthesis of polymer with polydisperse hard block, pTHF1100-U2PD.

=3325 cm-1 (strong)

'Free' urea group

Hydrogen bonded urea group

Synthesis and characterization of block copoly(ether urea)s