Thermodynamics and Phase Behavior

Since the block copolymer order-disorder transition (ODT) represents a balance between enthalpic and entropic interactions, measuring the ODT temperature (TODT) provides a means to quantify the strength of those enthalpic interactions (Flory-Huggins interaction parameter χ). With a database of such χ values, polymer-polymer miscibility—a key factor in any application—can be predicted. But since χ is a joint property of the two polymers, such a database is intrinsically quite large (of order ~n2, where n is the number of polymers)—far too large to measure in practice. If χ could be predicted from pure-component properties, this would reduce the required database to order ~n, making materials design much more tractable. The simplest theory is regular mixing, wherein each polymer is assigned a value of the Hildebrand solubility parameter. Recently, we have found that regular mixing is closely obeyed in all hydrogenated derivatives of styrene-isoprene copolymers, including “block-random” copolymers wherein one or more blocks is itself a random copolymer. This allows a decoupling of TODT from the molecular weight, and hence the block copolymer domain spacing; by reducing the compositional difference between the two blocks, which may be done continuously in the block-random architecture, the molecular weight and domain spacing at a constant TODT can be made arbitrarily large, as shown below for a range of near-symmetric block-random copolymers. Currently, we are exploiting this same block-random architecture to create polymers which have a very low χ against polyethylene, and will thus show miscibility in the melt.

Domain Spacing
Domain spacing (d) vs. number-average molecular weight (Mn) for two series of block-random copolymers of styrene and isoprene, hydrogenated using different catalysts to either retain the styrene aromatic unsaturation (SrhI), or convert the phenyl rings to vinylcyclohexane (VCHrhI). All polymers have thermally accessible TODT = 130 ± 50°C. Lines represent theoretical prediction of d ~ M2/3 in the strong segregation limit, and capture the data quite well; different lines are required for the SrhI and VCHrhI series because of their different statistical segment lengths. The red-circled points show that d can be practically tuned over more than a factor-of-5 range, at essentially constant TODT, by incorporating progressively less styrene (S) into the random block in the I-(SrI) precursors; both have a hydrogenated isoprene block (hI), but S-hI has a 100% S block, while hI-(SrhI)24 has a random block containing only 24 wt% S (Beckingham et al., Macromolecules, 46, 3084 (2013)).

Another way to increase the domain spacing (d) is to introduce polydispersity into the blocks. In joint work with researchers from Dow Chemical, we have investigated the structure of AB diblock and (AB)n multiblock copolymers prepared by a novel “chain-shuttling” polymerization, in continuous-flow stirred-tank reactors. A and B are both ethylene-octene random copolymers, but with very different octene contents; these polymers are thus also block-random copolymers. When such polydisperse block copolymers are in the weakly-segregated regime, chains with a short A block tend to dissolve in the B microdomains, and vice versa, greatly swelling the domain structure. Synchrotron SAXS data taken at the Advanced Photon Source on a flow-aligned specimen of one such diblock copolymer (below) shows an enormous d-spacing of 123 nm, even though the diblock molecular weight is only Mn = 69 kg/mol. This huge d-spacing imbues these materials with structural color: they appear blue in reflection and orange in transmission, and have been dubbed “photonic polyethylene”.

Polydispersity of Dibock Copolymers
Top left: cartoon of polydisperse diblock copolymers, where each block has the most-probable distribution, and the overall volume fractions of the black and orange blocks are each about 1/2. Bottom left: schematic of lamellar domain structure formed by these weakly-segregated olefin block copolymers, after flow-alignment in a lubricated channel die; the lamellar normals are predominantly oriented along the loading direction (LD). Right inset: two-dimensional SAXS pattern with the x-ray beam pointed along the constraint direction (CD). Three orders of lamellar reflection, with d = 123 nm, are visible. Right main panel: one-dimensional SAXS traces extracted from the two-dimensional pattern in the inset, either along the loading direction (LD, blue), showing three orders of lamellar reflection, or along the flow direction (FD, black), showing only very weak domain scattering in this highly aligned specimen (Li et al., Macromolecules, 43, 4761 (2010)).

Supported by the National Science Foundation, Polymers Program

Current/Recent Group Members, and Their Project Titles:

Will Mulhearn PhD *18 – “Melt-Miscibility in Block Copolymers Containing Polyethylene”
Adam Burns PhD *17 – “Thermoplastic Elastomers with Composite Crystalline-Glassy Hard Domains via Crystallization from a Single-Phase Melt”
Bryan Beckingham PhD *13 – “Mixing Thermodynamics of Block-Random Copolymers”
Sheng Li PhD *13 – “Structure and Properties of Novel Homopolymers and Block Copolymers Synthesized by Ring-Opening Metathesis Polymerization or Chain Shuttling Polymerization”

Selected Recent Publications

B.S. Beckingham and R.A. Register, "Regular Mixing Thermodynamics of Hydrogenated Styrene-Isoprene Block-Random Copolymers", Macromolecules, 46, 3084-3091 (2013).

B.S. Beckingham, A.B. Burns, and R.A. Register, "Mixing Thermodynamics of Ternary Block-Random Copolymers Containing a Polyethylene Block", Macromolecules, 46, 2760-2766 (2013).

S. Li, R.A. Register, J.D. Weinhold, and B.G. Landes, "Melt and Solid-State Structure of Polydisperse Polyolefin Multiblock Copolymers", Macromolecules, 45, 5773-5781 (2012).

S. Li, R.A. Register, B.G. Landes, P.D. Hustad, and J.D. Weinhold, "Crystallization in Ordered Polydisperse Polyolefin Diblock Copolymers", Macromolecules, 43, 4761-4770 (2010).