Understanding turbulent atomization through high-fidelity simulation and stability analysis of a two-phase mixing layer


Among the established fundamental problems in turbulent multiphase flows, one stands out. The problem associated with the mixing layer formed between liquid streams and parallel gases has recently attracted significant attention of researchers owing to its relevance in numerous industrial applications. In a recently published literature, it has been established that the instability of the gas-liquid interface due to the velocity difference between the two streams plays an essential role in the breakup of bulk liquids. Specifically, at the continuum level, the immiscible nature of the gas and liquid streams contributes to the atomization process in which the liquid breaks into small droplets thereafter mixing with the gas as a spray.

Unfortunately, the liquid breakups taking place at different spatial scales is attributed to the complexity and instability of the mixing layer. Despite several studies about multiphase turbulence in mixing layers, spray formation in a two-phase mixing layer between liquid and parallel gas streams have not been fully explored.

To this note, a recent international collaboration: Dr. Yue Ling (Baylor University), Dr. Daniel Fuster (Sorbonne Université, France), Professor Stephane Zaleski (Sorbonne Université, France) together with Professor Gretar Tryggvason (Johns Hopkins University) cross-examined two-phase mixing layers between the parallel gas and liquid streams. Specifically, they investigated the multiphase turbulence with the aim to obtain high order statistics useful in determining the influence of the upstream stability on the turbulence. Their work is currently published in Journal of Fluid Mechanics.

The research team started by conducting direct numerical simulation of the two-phase layer between gas and liquid streams. Next, the response of the multiphase turbulence flows was determined statistically by running the aforementioned simulations for a longer time. Furthermore, a detailed verification study was performed with four different meshes (the finest mesh consists of about four billion cells) to make sure spatially and statistically converged high order statistics in the turbulence are obtained and their significance on turbulent atomization are examined. To actualize their study, turbulent dissipation was evaluated, utilized to estimate the Kolmogorov and Hinze scales and then compared to the results obtained using the finest mesh.

The authors observed a dominant frequency during the simulations which was similarly observed in the temporal evolution of turbulent fluctuations. Consequently, the finest mesh enabled correct predictions of the Kolmogorov and Hinze scales with the estimated Kolmogorov scale being similar to the cell sizes. However, the estimated Hinze scale was by far larger than the droplets sizes and thus unsuitable for describing droplets formed through atomization, indicating that this conventionally-used scale is not a good measure of droplets size in atomization. In addition, the absolute nature of the instability of two-phase mixing layer was confirmed through the viscous stability analysis.

In summary, Dr. Yue Ling and colleagues successfully presented the multiphase turbulence statistics as well as the influence of the corresponding interfacial stability. The reported study is important because it addressed the fundamental problems in turbulent multiphase flows and thus a promising solution to many of the industrial applications such as fuel injection systems. 

About the author

Dr. Yue (Stanley) Ling is currently an Assistant Professor in the Department of Mechanical Engineering at Baylor University, Waco, TX. His current research focuses on simulation and modeling of multiphase flows, including atomization and sprays, droplets, microfluidics, and explosive dispersion of particles.

Dr. Ling obtained his PhD degree from University of Florida in 2010. Before joining Baylor, he was a postdoctoral researcher at University of Florida and later at Sorbonne Université, Paris, France.

About the author

Dr. Daniel Fuster is current CNRS Researcher at Institut Jean Le Rond D’Alembert, Paris, France. His research focuses on the development of numerical methods and models for the simulation of multiphase flows including cavitation and atomization. Dr Fuster received his PhD from University of Zaragoza, Spain, in 2007.


About the author

Dr. Gretar Tryggvason is the Charles A. Miller, Jr. Distinguished Professor in Mechanical Engineering and head of the Department of Mechanical Engineering at Johns Hopkins University. From 2002 to 2015, he served as Editor-in-chief of the Journal of Computational Physics. He is widely recognized for his contributions to computational fluid dynamics, including to the development of methods for computations of multiphase flows, as well as for pioneering direct numerical simulations of such flows. Dr. Tryggvason received his PhD from Brown University in 1985.

About the author

Dr. Stéphane Zaleski is Professor of Mechanics at Sorbonne Université (also known as University Pierre et Marie Curie or University of Paris 6). He investigates various numerical methods for the simulation of multiphase flow with applications for atomization, porous media flow and droplet impact. He is Associate editor of the Journal of Computational Physics and serves on the editorial board of several other journals. Dr Zaleski received his PhD at the Physics Department of Ecole Normale Superieure, Paris, France.


Ling, Y., Fuster, D., Tryggvason, G., & Zaleski, S. (2018). A two-phase mixing layer between parallel gas and liquid streams: multiphase turbulence statistics and influence of interfacial instability. Journal of Fluid Mechanics, 859, 268-307.

Go To Journal of Fluid Mechanics

Check Also

Flow separation downstream of a hydrofoil with a blunt trailing edge (plotted with zero angles of attack, q_in = 1 m/s at far upstream, Re = 1.0 × 10^6 ; the lengths are non-dimensionalized by the hydrofoil chord length c. - Advances in Engineering

A flow separation model for hydrofoil, propeller and duct sections with blunt trailing edges