46a Thermoplastic Elastomers Containing Crystalline and Glassy Components from Single-Phase Melts

John P. Bishop and Richard A. Register. Chemical Engineering, Princeton University, Princeton, NJ 08544

Introduction

Thermoplastic elastomers (TPEs) are typically symmetric ABA triblock copolymers made up of a majority component rubbery midblock (B) and minority component glassy endblocks (A) [1]. In this system, interblock repulsion between the hard glassy segments and the rubbery midblocks causes microphase separation, creating a physical crosslinking of the rubbery midblock chains, thus producing elastomeric behavior. In these types of TPEs where both the hard and soft blocks are amorphous, it is desirable to have a strong degree of incompatibility between the blocks so microphase separation can occur. However, the resulting microphase separated melts have an adverse effect on the processability of the material as the segregated melts have a high viscosity. If instead the hard segments were crystalline, microphase separation can be driven by crystallization itself from a lower viscosity single-phase melt and therefore strong interblock repulsion is not required. The use of crystalline endblocks in TPEs can also confer properties such as solvent resistance to the material [2]. Indeed, there has been a considerable amount of research on TPEs from single-phase melts containing crystalline endblocks, primarily using hydrogenated high-1,4 polybutadiene as the crystalline component [2-4]. Unfortunately, these crystalline TPEs were plagued by poor tensile strengths and high hysteresis relative to their amorphous counterparts.

Here, we report the synthesis of a TPE formed from a single-phase melt containing both crystalline and glassy hard segments. The idea here was to combine the superior tensile strength of a glassy block with the solvent resistance of a crystalline block, and allow microphase separation to be driven solely by crystallization from a single-phase melt. In order to accomplish this, we synthesized a symmetric pentablock copolymer with the architecture crystalline-glassy-rubbery-glassy-crystalline. With this architecture and appropriate selection of block lengths, crystallization from a single-phase melt will cause a glassy block-rich layer to form around the crystalline lamellae, forming an effective hard block with both crystalline and glassy components.

Experimental

We use living ring-opening metathesis polymerization (ROMP) and subsequent hydrogenation to synthesize our TPEs. This synthesis method provides access to model crystalline, glassy, and rubbery blocks suitable for incorporation into TPEs. These include hydrogenated polynorbornene (hPN), a highly crystalline polymer with an equilibrium melting temperature of 156°C [5]; hydrogenated poly(5-hexylnorbornene) (hPHN), a rubbery amorphous polymer with T_g= -22°C [6]; and hydrogenated polymethyltetracyclododecene (hPMTD), a glassy polymer with T_g = 163°C. ROMP polymerizations were conducted in a nitrogen atmosphere and involved the addition of a weighed amount of a Schrock-type initiator to toluene (previously purified by vacuum distillation from sodium benzophenone ketyl), followed by the addition of monomer charges (previously purified by vacuum distillation from metallic sodium). For the synthesis of triblock copolymers, the polymers were synthesized by sequential addition of monomer charges, and the reaction was terminated with a hundred-fold molar excess of benzaldehyde. For the synthesis of the pentablock, half of the polymer was synthesized by sequential monomer addition, which was followed by coupling the living ends together by terminating with a stochiometric amount of isophthalaldehyde to form the desired symmetric pentablock [7]. Hydrogenations were carried out in a 2 L Parr reactor at 500 psig H₂ and 100°C using Pd⁰/CaCO₃ as the catalyst.

Gel permeation chromatography (GPC) in THF was used to characterize each polymer's molecular weight distribution. Differential scanning calorimetry (DSC) data were acquired with a Perkin-Elmer DSC-7, calibrated on heating with indium and tin, with a scan rate of 10°C/min. Small-angle X-ray scattering (SAXS) patterns were acquired with an Anton-Paar compact Kratky camera equipped with a hotstage and an M. Braun OED-50M position sensitive detector. Uniaxial tensile stress-strain curves were obtained with either an Instron Model 1122 or 5865 on ASTM D1708 dogbones stamped from molded sheets at room temperature and a crosshead speed of 2 in/min.

Results and Discussion

An hPN-hPMTD-hPHN-hPMTD-hPN pentablock with both glassy and crystalline hard blocks was synthesized with target molecular weights of 7.5-7.5-120-7.5-7.5 kg/mol. The target molecular weights were chosen so that the polymer would form a homogeneous melt in order for microphase separation to be driven by crystallization of the hPN block. For comparison to our pentablock TPE with crystalline and glassy hard segments, we also synthesized a triblock TPE with just crystalline endblocks formed from a single-phase melt (verified by SAXS and the observation that the polymer flows under its own weight when heated above its melting temperature [8]), hPN-hPHN-hPN with a target molecular weight of 10-80-10 kg/mol, and a triblock TPE with only glassy endblocks, hPMTD-hPHN-hPMTD with a target molecular weight of 10-80-10 kg/mol. For this TPE, interblock repulsion must drive the microphase separation necessary for physical crosslinking.

SAXS and DSC experiments were performed to probe the microphase behavior of the pentablock and confirm that microphase separation in the polymer is driven by crystallization. The room temperature SAXS pattern of the pentablock shows a broad first-order peak consistent with a crystallization-induced microstructure. The microstructure persists until ~117°C, which coincides with the crystalline melting peak of the polymer observed by DSC. At 117°C and above, the SAXS pattern is featureless indicating a single-phase melt. The pentablock was also observed to flow under its own weight when heated above its melting temperature, further indicating that the polymer forms a single-phase melt [8]. The crystalline melting peak of the pentablock is very small (~1 J/g compared to 65 J/g for hPN homopolymer [5]), perhaps due to the formation of an hPMTD-rich layer surrounding the crystalline lamellae vitrifying during the early stages of crystallization, thus impeding the growth of the crystalline lamellae. The low level of crystallinity results in poor solvent resistance as the polymer is easily dissolved at room temperature in cyclohexane.

Uniaxial tensile testing of the pentablock shows desirable strain hardening behavior typical of all-amorphous TPEs and, more importantly, does not show any yielding. This is in stark contrast to the hPN-hPHN-hPN TPE, which shows plastic deformation of the crystals at even moderate strains, yielding a polymer with poor elastomeric properties. The pentablock's tensile strength of 7.4 MPa is lower than the 13 MPa of the amorphous hPMTD-hPHN-hPMTD triblock, perhaps due to the low levels of crystallinity in the pentablock discussed previously. We are currently working on obtaining greater levels of crystallinity by reducing the T_g (and delaying vitrification) of the hPMTD-rich layer by reducing the molecular weight of the hPMTD block and by reducing the relative amount of hPMTD in the pentablock (thereby increasing the relative amount of crystalline hPN).

References

1. Legge, N.R. Rubber Chem. Technol. 1987, 60, G83.

2. Morton, M. Rubber Chem. Technol. 1983, 56, 1096.

3. Morton, M.; Lee, N.-C.; Terrill, E.R. In Elastomers and Rubber Elasticity; Mark J.E., Lal, J., Eds.; American Chemical Society: New York, 1982; p. 101.