My broad interests are in developing and applying theoretical and computational techniques to better understand the behavior of a wide variety of dynamically evolving two-phase flow systems. A primary motivation behind my research is the need to understand the formation of self-organized structures in such systems. Such systems can range in size from the macroscale (several tens of meters) down to the micro-nanoscale (millionths to billionths of a meter). My research focusses not only on developing measures that characterize structure formation in such systems but also on developing the appropriate theoretical framework to describe the dynamics of the structure formation. While structure formation is more usefully characterized through higher-order (two-point or multi-point) statistics, a challenge is to model the effect of the structure formation on the behavior of the system at the single-point level. At the macroscale, these studies will lead to the development of scale and structure-based models for drag, heat and mass transfer for DNS, LES and CFD calculations of two-phase reactive flows. At the micro and nanoscale, these studies help us better understand various complex systems whose behavior is primarily controlled by clustering and aggregate formation. In addition, these studies will provide insight into the factors that control self-assembly in a variety of non-equilibrium systems at the micro and nanoscales. To complement the theoretical studies, I use a variety of continuum level simulation techniques that include immersed boundary and level sets, and molecular dynamics techniques such as the hard-sphere event driven paradigm. I also have interests in coarse-grained mesoscale computational techniques such as Brownian dynamics.
The objective of my current research is to appraise existing models and develop new, predictive physical models for liquid jet atomization. Toward this end, I perform detailed numerical simulations of liquid jets in a quiescent or moving ambient (gas) using state-of-the-art interface tracking techniques. These numerical simulations provide a glimpse of the intricate structures in the near-field regions of the liquid jet that form precursors to ligament and drop formation. These simulations can help us accurately quantify near-field liquid and ambient gas statistics that are challenging and sometimes impossible to measure in experiments. Guided by physical modeling principles, we plan to use these statistics to propose predictive models for liquid primary break-up.
Detailed numerical simulations of primary breakup of liquid jets
LES of particle-laden turbulent flows
Working in the multiphase flow computations group at Iowa State introduced me to several unsolved problems and challenges, nuances and subtleties involved in the mathematical description and modeling of multiphase flows. Interestingly, this complexity in description and modeling is encountered even in (simple) homogeneous two-phase flows. Studies of homogeneous two-phase flows form an important and necessary first step towards understanding real-world, complex, inhomogeneous two-phase flows. Furthermore, there is evidence in literature of widely used two-phase flow models failing to be predictive in such simple two-phase flow configurations. Therefore, the focus of my work at ISU primarily revolved around building new predictive models in homogeneous two-phase flow configurations.
Development of consistent PDF formulations for multiphase flows
Multiscale interaction models for Lagrangian-Eulerian spray computations
Dual timescale Langevin model for particle-laden turbulent flows
Direct numerical simulations of particle-laden flows using immersed boundary methods
Molecular dynamics simulations of canonical granular flows
Accurate Numerical Implementation of Spray Vaporization models using Particle methods
