Heat Transfer Modification in a Particle-Laden Turbulent Channel Flow

Abstract

Heat transfer in multiphase flows plays an important role in many industrial applications. For instance, particle-based solar receivers utilize the high absorptivity and heat capacity of dispersed phase in a carrier fluid to improve efficiency and heat transfer. This dispersed phase generally consists of a large number of small particles. It is, therefore, difficult to completely resolve such flows considering their finite size. Usually, these particles are so small, that they can be treated as point particles. This considerably reduces the computational effort, while preserving the essential characteristics of the particle-laden flows. In this thesis, the focus is on heat transfer modulation in a particle-laden turbulent channel flow using direct numerical simulations.

In flows with temperature gradients (for example, channel flow between hot and cold wall), particles absorb heat from hotter regions and release heat to colder regions, thereby enhancing heat transfer through particle feedback flux. On the other hand, presence of particles leads to decay in turbulence resulting in lower turbulent heat transfer. The interplay of this two phenomena can either increase or decrease the overall heat transfer based on Stokes number and thermal Stokes number (ratio of thermal response time to characteristic time scale of the flow).

To investigate the heat transfer modulation in particle-laden channel flow, the existing DNS code developed by Boersma [5] has been modified to include particle transport and heat transfer. The point-particle approach with two way coupling is implemented using trilinear interpolation scheme and 3 rd order Runge-Kutta time marching scheme. The implemented code is validated using the results from literature [20] for a flow with no external heating.

With the developed code, cases with no external source term and external source term with different optical thickness of the fluid have been analyzed. In order to focus only on the fluid-particle interaction, the effect of gravity is neglected and the flow is considered to be incompressible. It has been observed from these simulations that particles play an important role in modulation of heat transfer in such flows. Mean temperature profiles, heat flux mechanisms, temperature variance and budgets of temperature variance are studied extensively in order to understand the underlying phenomenon. It is found that the particle feedback heat transfer is the dominating mode in particle-laden flows.