(289c) Thermoplastic Elastomers Via Crystallization from Homogeneous Melts

Burns, A. B., Princeton University
Register, R. A., Princeton University

The unique mechanical properties of elastomers are predicated on the formation of a nanoscopic network of crosslinks. In block copolymer thermoplastic elastomers (TPEs), such a network is produced by chemical incompatibility between the blocks. The prototypical TPE is a symmetric triblock copolymer having a rubbery midblock and glassy endblocks. With the appropriate choices of block lengths, microphase separation produces discrete glassy domains which physically crosslink the rubbery matrix.1 Despite the physical crosslinks being labile at elevated temperatures, the persistence of microphase separation into the melt leads to high viscosity and elasticity, making processing difficult. Additionally, these amorphous polymers have poor solvent resistance. Replacing the glassy endblocks with crystalline blocks can ameliorate both of these issues. By using crystallization to drive physical crosslinking, microphase separation can be avoided and homogeneous melts can be obtained.2–5 However, the extended plate-like crystals formed by the crystalline endblocks impart qualitatively different mechanical behavior–namely, the modulus increases and yielding is observed in conjunction with significant unrecoverable deformation.4,5

To leverage the advantages of both crystalline and glassy blocks, a short glassy block is inserted between the crystalline and rubbery blocks.5This block sequence is designed to permit access to single-phase melts, wherefrom physical crosslinking is achieved through crystallization of the endblocks. Due to the connectivity of the blocks, the glassy blocks are concentrated around the crystallites providing mechanical reinforcement, and ultimately conferring the desired elastomeric response under applied stress.

Materials having this block sequence are synthesized by sequential anionic polymerization of the butadiene, styrene, and isoprene followed by coupling to produce a symmetric pentablock copolymer. Although anionic polymerization of these common monomers is well-established, several challenges had to be addressed. To minimize vinyl addition in the polybutadiene block (thus maximizing the crystallinity eventual polyethylene block after hydrogenation) the butadiene is polymerized in cyclohexane. However, the crossover from butadiene to styrene is slow in aliphatic hydrocarbons, producing a broad molecular weight distribution in the short polystyrene block. Adding benzene (50% v/v) prior to the styrene block polymerization accelerates the crossover, yielding a narrow molecular weight distribution without significantly altering the microstructure of the ensuing polyisoprene block. Finally, after polymerizing the isoprene block, the living triblocks are coupled with dimethyldichlorosilane to produce pentablocks. Small amounts of tetrahydrofuran are added to reduce the reaction time for coupling from days (in apolar solvents) to seconds, enabling near stoichiometric coupling via titration. Subsequent hydrogenation is used to convert the low-vinyl polybutadiene block to semicrystalline polyethylene and improve the overall thermal stability.

Through careful choices of block lengths single-phase melts are achieved, as demonstrated by small-angle x-ray scattering (SAXS). Consequently, the steady-shear viscosity of the pentablocks at 180 °C is 30 to 300 times less than that of a microphase separated triblock with glassy endblocks, prepared analogously. Below the melting point, the SAXS data reveal two characteristic peaks. The larger real-space length scale is taken to be the average spacing between physical crosslinking domains (ca. 60nm), while the smaller length scale coincides with the crystal-amorphous repeat distance of hydrogenated polybutadiene. Atomic force microscopy reveals hard domains dispersed in a rubbery matrix with poor long-range order. This solid-state structure confers the characteristic mechanical behavior of an elastomer. Increasing the glassy block fraction improves the performance by reducing the Young’s modulus and increasing the ultimate strength. When the endblock crystallinity is low, as is the case in this work, the improvements are modest due to the already limited size and connectivity of the crystallites. However, in the case of high crystallinity hydrogenated polynorbornene endblocks studied by Bishop and Register5the glassy block has a dramatic impact on the mechanical properties by greatly reducing the crystallinity and crystal size.

Despite these improvements, the strain recovery of the pentablocks tracks that of the triblock with crystalline endblocks. Star block copolymers, where each arm is a block copolymer with the crystalline-glassy-rubbery structural motif have been investigated as a way of improving the recovery. These polymers are synthesized analogously to the linear materials where the functionality of the chlorosilane coupling agent is greater than two. It is hypothesized that the permanent crosslink at the center of each star can further improve the mechanical performance, without adversely affecting the phase behavior. However, it is found that the potential advantages are overcome by the difficulty in preserving the star architecture through the hydrogenation step.

This work was generously supported by the National Science Foundation, Polymers Program (DMR-1402180).


  1. Pakula, T.; Saijo, K.; Kawai, H.; Hashimoto, T. Macromolecules 1985, 18, 1294–1302.
  2. Morton, M.; Lee, N.-C.; Terrill, E. R. Elastomeric Polydiene ABA Triblock Copolymers with Crystalline End Blocks. In Elastomers and Rubber Elasticity; Mark, J. E.; Lal, J., Eds.; ACS Symposium Series 193; American Chemical Society: New York, NY, 1981; pp 101-118.
  3. Koo, C. M.; Wu, L.; Lim, L. S.; Mahanthappa, M. K.; Hillmyer, M. A.; Bates F. S. Macromolecules 2005, 38, 6090–6098.
  4. Myers, S. B.; Register, R. A. Macromolecules 2009, 42, 6665–6670.
  5. Bishop, J. P.; Register, R. A. Macromolecules 2010, 43, 4954–4960.