Flow-based polymer processing is ubiquitous in the formation of soft materials for applications ranging from healthcare to packaging. Polymer processing strategies have historically been determined empirically, where the maximum flow rate is often established by the onset of visual distortions from uniform flow. Such an approach prohibits predictive flow calculations for arbitrary flow environments and polymeric formulations. Theoretical advances over the past several decades have enabled molecularly informed rheological modeling of entangled polymer liquids; however, most models assume that the fluid composition remains uniform during flow. Several exceptions to this assumption are known for polymer solutions such as shear-enhanced concentration fluctuations and polymer migration across curved streamlines, but these effects are typically not accounted for in models due to the belief that these compositional heterogeneities either occur on a small scale, with an insignificant effect on the bulk flow behavior, or on diffusive timescales, which are much longer than typical processing flows. This belief has been challenged by a recently developed two-fluid theory for entangled polymers, which, under certain conditions, predicts that uniform flow is unstable due to a shear-induced demixing instability that results in shear banding (i.e., two or more regions of distinct shear rates across the fluid under a shear flow); however, there is a lack of experimental measurements to directly compare with this theory.

To fill this gap, this dissertation experimentally explores the coupling of shear-induced compositional changes to bulk flow behavior in entangled polymer liquids. To do this, a combination of novel rheo-optical techniques are developed and used to quantify both fluid velocity and concentration fields. Qualitatively different results are observed depending on the number of entanglements and osmotic susceptibility of the polymer solutions. For entangled polystyrene in dioctyl phthalate, spatial variations of shear-enhanced concentration fluctuations in startup shear are coupled to changes in the bulk flow. In particular, the presence of mesoscopic concentration heterogeneities coincides with a local effective viscosity increase of the fluid. Conversely, entangled polybutadiene in dioctyl phthalate exhibits macroscopic heterogeneities in concentration that coincide with shear banded velocity profiles, consistent with a two-fluid model that couples the flow to polymer concentration. The effect of polydispersity on the flow behavior of entangled polymer liquids was investigated by using chemically homologous, bidisperse polymer blends. Interestingly, differences in the flow behavior were observed for blends of equivalent numbers of entanglements, but different blend compositions. Lastly, the two-fluid model for entangled polymer solutions is used to investigate nonuniform flows that arise from shear-induced demixing in complex flow protocols (i.e., large amplitude oscillatory shear (LAOS)).

The results of this dissertation underscore the possibility for non-local concentration changes in flows of entangled polymeric liquids, and reveal a wide range of flow phenomena that result when it is operative. More generally, the concept of flow-concentration coupling is likely applicable to many other complex fluids and should be accounted for in future models to accurately describe their rheology.

Flow processing of polymers and wormlike micelles usually involves nonlinear deformations, which can significantly modify both the associated microstructural configuration and dynamics. Determining the connection between processing, structure, and properties remains a grand challenge due to limitations in currently available tools. Thus, the primary focus of this dissertation is to develop combined experimental, theoretical, and computational approaches to gain a deeper understanding of the processing-structure-property relationship of polymers and wormlike micelles.

Wormlike micelles (WLMs) are long, semi-flexible chainlike structures formed by the self-assembly of surfactants and are ubiquitously used in oil and gas industry, as well as in consumer products. The rheology of wormlike micelles is critical to the successful formulation and engineering of these products and processes. Although it is widely accepted that equilibrium micelle scission dynamics greatly influences the rheology of WLMs, there is still considerable theoretical debate regarding whether scission dynamics is affected by flow under nonlinear deformations. Direct structural measurements in flow are needed to directly answer whether and how flow affects scission of WLMs.

In situ small angle neutron scattering (SANS) represents a powerful technique for measuring material microstructures under flow. However, SANS methodology for studying wormlike micelles and polymers is currently limited in terms of available nonequilibrium scattering models and in terms of experimental analysis methods to deconvolute effects of chain orientation, stretching, inter-chain interactions, and changes in chain length. With respect to scattering models in flow, this dissertation develops a connected-rod model for semiflexible chains in flow and achieves excellent agreement with experimental anisotropic scattering results from wormlike micelles. We also formulate a scattering model for dilute, flexible polymers in shear flow and use results from Brownian dynamics simulations for the polymer conformation in flow. To address the question of how and whether flow affects scission of WLMs, we conducted systematic flow-SANS and rheology experiments on a series of linear wormlike micelles. A combination of SANS modeling, steady-state flow-SANS experiments, and time-resolved flow-SANS experiments enables direct microstructural measurement of wormlike micelles in flow and strongly suggests the presence of flow-enhanced scission.

Additionally, for relating microstructural information to macroscopic dynamic properties of the material, a recently advanced rheometry technique, orthogonal superposition (OSR), is predicted to be very useful. However, relatively little is known about how to interpret the nonlinear viscoelastic results in the context of entangled polymer dynamics. Specifically, there is a need for a deeper theoretical and computational study to provide a fundamental basis for interpreting OSR measurements. We combine numerical calculations and a perturbation analysis using detailed microstructural models to study orthogonal superposition for monodisperse and polydisperse entangled linear polymers. We find that orthogonal superposition gives very useful information about nonlinear material moduli under flow, which can provide better sensitivity for testing constitutive models for nonlinear polymer processing. Results in our work have important implications for the design and interpretation of future OSR experiments.

The tools we develop in this dissertation are important for understanding the rheology, scattering, and microstructures of not only wormlike micelles and polymers, but also other complex fluids that share similar underlying physics.