Crystallizable Block Copolymers
Although commercial block copolymers used as melt-processible rubbers and adhesives are based on amorphous blocks, designing crystallinity into one of the blocks can confer significant property advantages, such as vastly improved solvent resistance. Combining two or more self-organizing mechanisms, such as crystallization and interblock repulsion, into a single polymer yields both morphological richness and kinetic complexity.
We have shown that crystallization can radically change the mesophase diagram from the usual "spheres-cylinders-gyroid-lamellar" progression:when crystallization occurs from a single-phase melt, a lamellar morphology results for a wide range of compositions. Time-resolved SAXS/WAXS/DSC (with Professor Tony Ryan, University of Sheffield) demonstrated that the structure in these materials, which extends hierarchically from Angstroms to microns (see figure below), develops rapidly and simultaneously on all length scales. Finally, these diblocks freely cocrystallize with the homopolymer of the crystallizable block, indicating their utility as compatibilizers for polymer blends. The material design guidelines in these systems are remarkably straightforward: the structure is built in through the molecular architecture, and is essentially decoupled from processing.
Morphological hierarchy exhibited by block copolymers crystallizing from single-phase melts. Left to right: spherulites; block copolymer microdomains; lamellar crystals; crystalline block unit cell. Morphological characterization techniques by which each level was established: small-angle light scattering (SALS); polarizing optical microscopy (POM); small-angle x-ray scattering (SAXS); wide-angle x-ray scattering (WAXS).
Having established the morphology of block copolymers crystallizing from single-phase melts, we examined whether microphase-separated melts could effectively template crystallization: for example, could crystallization be restricted to occur only within the cylinders of a hexagonal mesophase, even when the matrix polymer remains above its glass transition temperature? We found that while the structure present in weakly-segregated block copolymers was immediately destroyed upon crystallization, by increasing the interblock segregation strength, crystallization could be constrained to follow the pre-formed structure established in the melt (see figure below). A striking consequence of confined crystallization is that such materials remain optically transparent when crystallized, as no spherulites or other micron-scale structures--the top level of the structural hierarchy--are formed. Confined crystallization also effects a pronounced orientation of the crystalline block's unit cell--the bottom level of the structural hierarchy--when the microdomains are anisometric (cylinders or lamellae). These novel structures are stable even upon extended annealing just below the melting point, and thus reflect a robust means of morphological control.
Strongly-segregated diblock containing spherical microdomains of “polyethylene” (hydrogenated low-vinyl polybutadiene, denoted E) in a matrix of styrene-ethylene-butene terpolymer (SEB, rubbery: Tg < E block Tc). Left: transmission electron micrograph of fully-crystallized E/SEB 9/54 diblock showing crystalline E spheres (white) in a matrix of RuO4-stained SEB. Right: SAXS patterns for E/SEB 9/54 in the melt and fully crystallized. Retention of second-order peak and constancy of first-order peak position indicate that the melt morphology is preserved upon crystallization.
More subtly, by dividing the crystallizable material into discrete nanometer-scale domains, such as cylinders or spheres, one nucleation event per domain becomes necessary to crystallize the material. This is evident in both the structure of the material, where isolated crystals are observed (see figure below), and in the crystallization kinetics, which are conveniently tracked via time-resolved SAXS/WAXS/DSC. When crystallization is fully confined, the crystallization kinetics are first-order rather than the usual sigmoidal case, with the deep undercoolings and strong temperature dependence characteristic of homogeneous nucleation. By contrast, if the crystallizing domains become interconnected (either through an intrinsic feature of the morphology or through defects formed by crystallization), then sigmoidal crystallization kinetics are observed. In a collaboration with Professor Jamie Hobbs of the University of Sheffield, we have examined this “breakout” process in real space and real-time, via rapid-scan hot-stage atomic force microscopy.
Transmission electron micrograph of a flow-aligned block copolymer where crystallization is confined to cylinders. View is down the cylinder axis; matrix is poly(vinylcyclohexane). Elliptical cross-sections of “polyethylene” (hydrogenated low-vinyl polybutadiene, E) cylinders each contain a bright central stripe, which is the unstained E crystallite running along the length of the cylinder; each bright stripe is flanked by two dark stripes, which are RuO4-stained amorphous E layers extending to the microdomain wall. Note that exactly one crystal per cylinder is observed.
All of the above results were obtained on materials synthesized via “living” anionic polymerization, where the crystallizable block is hydrogenated low-vinyl polybutadiene (hPBD). hPBD resembles a linear low-density polyethylene: the ethyl branch defects which result from unavoidable 1,2 units in the polymer limit both the crystallinity (to <40%) and the crystal thickness (to ca. 5 nm average). More recently, we have extended our work into new defect-free crystallizable block copolymers synthesized by “living” ring-opening metathesis polymerization (ROMP). The crystallizable block can be either a perfectly linear polyethylene (prepared by hydrogenation of polycyclopentene), or hydrogenated polynorbornene (hPN).
When these highly-crystalline blocks are paired with monomers yielding single-phase melts, then the lengths of the amorphous blocks can be used to precisely control the crystal thickness. While control of the microdomain dimensions (e.g., lamellar thickness) is now well-established in amorphous polymers, it has not been demonstrated previously for synthetic crystalline polymers; rather, the crystal thickness is usually set kinetically, according to the material’s thermal history. But this idea of “encoding” a particular folding structure into the primary macromolecular architecture is standard dogma in complex biological macromolecules. The idea is shown in the figure below, where attaching a progressively longer amorphous (red) block to the crystallizable (blue) block induces it to fold a progressively greater number of times. This effect was predicted by DiMarzio, Guttman, and Hoffman (DGH) in 1980, who also presented the theoretical scaling laws governing the crystal-amorphous periodicity (d) as functions of the total chain (Nt) and amorphous block (Na) lengths. As the figure below shows, we have confirmed this prediction using diblock copolymers of hPN and hydrogenated poly(ethylidene norbornene), hPEN. As anticipated, reductions in crystal thickness translate directly into reductions in melting point, allowing us to engineer polymers having a narrow, tunable, and predetermined melting range.
Scaling plot for hPN-hPEN (red-blue) diblocks. Data points represent equilibrium crystal-amorphous repeat distances determined by SAXS; the x-axis is the log of the amorphous block length, Na. Cartoons represent the number of folds in the crystalline block (n) as the amorphous block is lengthened. The DGH scaling prediction is shown in the upper right corner, and closely matches the experimental data.
Supported by the National Science Foundation, Polymers Program
Recent/Current People and Projects:
Sasha Myers GS – Polyethylene Block Copolymers
John Bishop GS – Structuring Polymer Crystals through Macromolecular Architecture
Li-Bong Lee *04 – Polymer Crystalline Texture Controlled Through Block Copolymerization
Lynn Loo *01 – Controlled Polymer Crystallization Through Block Copolymer Self-Assembly
Selected Recent Publications:
J.K. Hobbs and R.A. Register, “Imaging Block Copolymer Crystallization in Real Time with the Atomic Force Microscope”, Macromolecules, 39, 703-710 (2006).
L.-B.W. Lee and R.A. Register, “Equilibrium Control of Crystal Thickness and Melting Point through Block Copolymerization”, Macromolecules, 37, 7278-7284 (2004).
Y.-L. Loo, and R.A. Register, “Crystallization Within Block Copolymer Mesophases”, Chapter 6 in Developments in Block Copolymer Science and Technology, I.W. Hamley, ed. (Chichester: John Wiley & Sons, 2004), pp. 213-243.