Dense gases, characterized by complex molecules and by a highly non-‐ideal thermodynamic behaviour, are of interest for many engineering applications. The most important one is represented by energy conversion cycles, like Organic Rankine cycles (ORC) and heat pumps. Progress in these applications can be achieved by developing advanced design models and tools, while experimental investigations remain extremely difficult and scarce.
The more striking peculiarity of dense gases is that they may exhibit non-‐classical gas dynamic phenomena in the transonic and supersonic regime. Additionally, their heat capacity is generally much higher than standard gases like air, leading to the decoupling of thermal and dynamic effects. Finally, the transport properties of dense gases exhibit a very different behaviour with respect to perfect gases : for instance their dependency on the density (or pressure) is no longer negligible and the dependency on temperature is somewhere in between those of liquids and gases. All these properties affect the global statistics and the small-‐scale mechanisms of turbulence. For these reasons, we are conducting a thorough study of turbulent dense gas flows by means of Direct Numerical Simulations, in order to understand how and to which extent dense gas effects modify turbulence compared to standard gases and to assess the validity of turbulence models for this kind of flows.
The present talk presents some recent studies in dense gas turbulence. On the one hand, we consider the decay of homogeneous isotropic turbulence at high turbulent Mach number and we investigate how dense gas effects affect the flow topology and the dissipation and enstrophy generation mechanisms. On the other hand, DNS of a supersonic turbulent channel flow show that dense gas effects modify both the global flow properties (as the skin friction and the heat flux) and the statistics of thermodynamic quantities, while the statistics and typical structures of the dynamic field remain rather close to those of an incompressible flow.
In the perspectives, the work in progress to extend our investigations to more complex geometries, to quantify uncertainty bounds of the simulated flow fields and, in the end, to improve design tools for ORC turbines will be also briefly discussed.
Contact LMFA : Alexis Giauque