Turbulence is the most common type of fluid flow in both the industry and the environment, while it is still one of the unsolved problems in physics. Its ability to mix and disperse energy or mass (like chemicals or particles) plays a key role in every day applications, like flows in oil and gas pipelines, in industrial reactors, in mixing, in heating or cooling, as well as in the atmosphere and in oceans. The goal of this project is to understand the effects of flow features that are commonly observed in turbulent flow on the transport of energy and mass using advanced computational techniques. The findings of this research could lead to efficient energy management, efficient design of industrial equipment where heat and mass transfer take place, and better prediction and control of pollutant dispersion. The database generated through this project will become available to the scientific community, and simulation data generated through this project will be used to develop animations of turbulent dispersion that can be used in K-12 science demonstrations.
Central concerns for the development of a comprehensive turbulent transport theory have been the prediction of the spatial variation of turbulent viscosity, the effect of molecular dispersion on turbulent transport, and the effect of coherent velocity structures on the transport properties. The main hypothesis here is that the reason for the failure in predicting scalar transfer from momentum transfer is that only some of the velocity structures participate in turbulent transfer close to a wall, and the range of scales that participate in the transfer depends on molecular dispersion effects. Main questions to be answered are what is the role the very large scales of motion (known as VLSM) in turbulent transport, and how the interplay between molecular diffusion and convection affects mixing or separation of particles in anisotropic turbulence. The proposed approach is to use computations with Lagrangian methods for analysis, which afford the study of a range of fluids that span several orders of Prandtl number in magnitude (e.g., liquid metals, gases, refrigerants, and electrochemical fluids) making it possible to handle cases where conventional methods are often not feasible with the current supercomputers. Expected results could lead to the development of a comprehensive model for the prediction and control of turbulent transport.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
