September
16
2025
11:00 AM

Location

Guggenheim 442
Tuesday, September 16, 2025 11:00AM

You are invited to attend

AE Seminar: Makrand Khanwale

 

 

"Unravelling the complexities of breakup dynamics in turbulent multiphase flows through high-fidelity simulations"

Tues., September 16

11:00 a.m.

Guggenheim 442

Abstract:

One of the long-standing challenges in turbulent multiphase flows that involve deforming interfaces is predictive modelling of the process of interface breaking apart. This is a multiscale phenomenon that involves effects even from molecular scales. Understanding and predicting this process is crucial for many manmade and natural processes like atomization, disease spread, energy conversion processes, inkjet printing, biological flows, and wave breaking in oceans. "High-fidelity" numerical simulations can be valuable in understanding such complex multiscale phenomena.  In this talk, I will discuss the formulation and numerical discretization of phase-field models to model and simulate such multiphase flows involving complex breakup phenomena. I will detail the advances in numerical analysis to maintain stability and accuracy of discretization and computational strategies that enable unprecedented scale resolution while keeping computational costs low, including deployment of the discretization on a massively parallel adaptive meshing framework.  I will use test problems of progressively increasing complexity in both low and high Reynolds number regimes to demonstrate the effects of modelling and computational approaches. I will then present simulations and analysis of the canonical pulsed primary jet atomization problem that demonstrates the capabilities of the computational framework. Through this case, I will discuss the intricacies of the dynamics of atomization, which involves a complex cascade of instabilities and break-up mechanisms involving sheet rupture to filament formation, leading to droplet pinch-off.  To quantitatively understand the effect of our approach on the breakup statistics, the droplet size distribution will be analysed and put in context with the other current state-of-the-art modelling approaches. Further, I will outline how my work lends itself to extension to other multiphysics phenomena, like heat and mass transfer, charge transport in electrolysis systems, non-Newtonian multiphase flows, and compressible multiphase flows. I will also present some recent work on modeling turbulent flows using direct numerical simulations of carefully designed canonical settings. Additionally, I will also present my overarching research vision and plans. 

Bio: 
 
Makrand Khanwale is currently a Physical Science Research Scientist at the Department of Mechanical Engineering at Stanford University, working with Prof. Ali Mani and Prof. Gianluca Iaccarino. As a part of his work at Stanford, he is involved in multiple projects involving turbulence modelling, compressible multiphase flows, and multiphase electrochemical systems. Makrand received his PhD at Iowa State University, co-majoring in Mechanical Engineering and Applied Mathematics. He was co-advised by Dr. Baskar Ganapathysubramanian and Dr. James Rossmanith. As part of his PhD, he developed energy-stable numerical methods to simulate two-phase flows using Cahn-Hilliard Navier-Stokes equations. Makrand won both the Research and Teaching Excellence awards during his time at Iowa State University. He also has experience in the development of tools to analyse and understand complex physical processes like multi-phase flows and turbulence. Before joining Iowa State for his graduate work, he had a brief stint as a Research Associate in Dr. Krishnaswamy Nandakumar‘s group at Louisiana State University. There, he worked on analytical theories of energy transfer processes in turbulent multi-phase flows. Makrand earned his Bachelors degree in Chemical Technology from the Institute of Chemical Technology, Mumbai. Makrand's research interests lie at the intersection of multiphysics fluid mechanics, turbulence, numerical analysis, and scalable scientific computing.