The project is devoted to the modeling and numerical simulation of yield stress fluids, and studies of viscoelasticity and confinement in the flow of two immiscible fluids. The majority of the effort is focused on the following three themes: (i) Recent developments have enabled detailed measurements of the properties of a class of suspensions known as thixotropic yield stress fluids. An example is the flow of ketchup under applied stresses. Prior theoretical models require the input of measured quantities such as a yield stress, or to pose a phenomenological equation for the structure of the microcomponents. In contrast, a new mathematical perspective is explored in this project, in which yielding is an output of a systematically derived viscoelastic constitutive model in the limit of large relaxation time. Direct numerical simulations, in conjunction with multi-scale perturbation techniques based on slow and fast time scales, will enable the prediction of several experimentally observed phenomena, such as the hysteretic loop for up-down ramping of applied stress. (ii) The technological drive toward smaller scales has spawned new modeling and simulation challenges for droplet evolution in devices that are as small as the droplet itself. Confinement can enhance drop elongation, lead to new modes of drop breakup, and introduce numerical difficulties with respect to the constitutive laws and interface tracking. Capillary-focusing is a recently developed technology for the production of monodisperse submicron-sized droplets, which is not completely understood. This project sheds light on the physical mechanisms at work, by modeling the Hele-Shaw geometry and with direct numerical simulations. (iii) The advantages in combining the desired properties of two different liquids to produce a new blended mixture are well known. The quality of the blend is strongly dependent on the size of the dispersed droplets and their distribution. Therefore, it is of practical importance to understand how a droplet evolves in a matrix fluid, and to predict the influence of the rheological properties of each constituent. A novel viscoelastic height function algorithm will be developed and implemented in this project, and optimized for drop breakup simulations with possibly high gradients of stress.
This project contributes to the understanding of physical mechanisms that control practical applications, such as recycling plastics for environmental sustainability, predicting the dynamics of droplets and particulates in confined flows for biomedical drug delivery, and understanding fundamental processes that improve nanofluidic technology for the pharmaceutical industry. The use of a computational infrastructure such as XSEDE will enable direct numerical simulations for comparison with experimental data. The knowledge gained from this research will be disseminated to a broader scientific audience at conferences on mathematics, physics and engineering, and through journal publications. The outcomes include the training of graduate students, outreach to the local K-12 audience, and cross-disciplinary partnerships.
