We report the first clear observations of an entropy-driven phase transition between a dilute micellar “gas” and a disordered but highly associated micellar “liquid” realized with aqueous solutions of poly(ethylene oxide) chains fully end-capped with C16 and C18 hydrophobes. Dynamic light scattering and capillary viscometry determine the radii and aggregation numbers of the micelles and, together with the coexisting concentrations, permit estimates of the strength of the entropic attraction through the adhesive hard-sphere model. The behavior is qualitatively consistent with expectations for the entropy gain from the exchange of end blocks between cores of flowerlike micelles consisting of associative triblock copolymers.

We report the rheology of the micellar solutions formed from narrow-molecular-weight poly(ethylene oxide) chains fully endcapped with C16 or C18 alkanes and correlate the viscoelasticity, characterized by the high-frequency modulus G ∞‘ and a single relaxation time λ, with the measured characteristics of the micellar solutions. Scaling G ∞‘ by p 3/2 kT/τ R H 3 (p = aggregation number, R H = hydrodynamic radius, 1/τ = attractive virial coefficient, k = Boltzmann's constant, and T = temperature) and plotting against the hydrodynamic volume fraction of micelles collapses the two sets of data. The relaxation time scales as the diffusion time, μ R c 3/kT, for a free hydrophobe escaping from a micellar core of radius R c times a Boltzmann factor accounting for an association energy that increases linearly with hydrophobe length. The low shear viscosity follows as η o = λ G ∞‘.

We have used pyrene fluorescent probe experiments to study aqueous solutions of a poly(ethylene oxide) (PEO) of M = 35000 with C16H33 end groups. The end groups were attached by reaction of the PEO with hexadecyl isocyanate, and the polymer was then purified so that essentially all of the polymer in the sample contained 2.0 end groups/polymer molecule. The hydrophobic end groups of this polymer associate in water to form micelle-like structures which serve as solubilization sites for the pyrene molecules. Fluorescence decay profiles for both pyrene and 1-ethylpyrene were fitted to the Poisson quenching model to obtain the parameters n, the mean number of quenchers per micelle, and k, the pseudo-first-order rate constant for the quenching reaction in the micelle. While the data seem relatively well-behaved, we obtain different values for the end-group aggregation number (N R) for the two probes. For ethylpyrene, N R = 16 and k = 10.4 μs-1, while for pyrene, N R = 21 and k = 8.7 μs-1. Plots of n versus [Py] or [EtPy] are linear for various polymer concentrations, but exhibit a nonzero intercept in the limit of low quencher concentration. We examine a number of possible explanations for this behavior.

The characterization of hydrophobically modified alkali-soluble emulsions (HASE) using conventional techniques such as gel permeation chromatography or static light scattering is difficult because of the hydrophobic association. Two different approaches were taken to prevent the hydrophobic association of HASE polymers in aqueous solution: (1) hydrolyze the polymer to dislodge the hydrophobic constituents, and (2) use methyl-β-cyclodextrin, which has a hydrophobic cavity and a hydrophilic outer shell, to shield the hydrophobes from associating. Using these two approaches, the molecular weight (M w), hydrodynamic radius (Rh), and radius of gyration (Rg) of single chains of these polymers were determined using gel permeation chromatography (GPC) and dynamic (DLS) and static (SLS) light scattering techniques. The molecular weight of the control polymer (i.e., polymers with similar composition but with no hydrophobe) is found to be ∼7.0 × 10 5 g/mol. The molecular weights of these polymers with C8, C16, and C20 hydrophobes were found to be ∼2.0 × 10 5 g/mol. By comparing GPC and SLS results, we were able to determine that, except for one chemical site, branching or grafting did not occur in the polymer chain during synthesis.

No abstract