Algebraic zero-equation versus complex two-equation turbulence modeling in supercritical fluid flows

Heat transfer data is essential to improve the design of Super-Critical Water Oxidation (SCWO) systems. There are large discrepancies among the results of numerical studies of heat transfer in a supercritical environment especially for high heat flux conditions. The difficulty in turbulence modeling is considered to be one of the main sources of such discrepancies. Investigating the turbulence modeling in a vertical pipe flow of supercritical water is the main objective of the present study. A good number of various turbulence models, from algebraic zero-equation to complicated two-equation low Reynolds number k–ε models, have been incorporated in a 2D numerical code. The results appear to be quite sensitive to the choice of the turbulence model, especially at flow conditions leading to heat transfer deterioration. In addition to predicting the flow behavior more accurately, a proper choice of turbulence model in a supercritical fluid flow may effectively improve the computing efficiency of the numerical code. The turbulence model which leads to the best agreement of the results of present numerical study with the experiments has been determined for both cases of the enhanced and deteriorated heat transfer. The effect of turbulence anisotropy in supercritical fluid flows has also been investigated.

► Various types of turbulence models in supercritical fluid flows have been examined.
► When buoyancy is weak, algebraic models are more efficient than complex models.
► Algebraic models cannot capture the flow behavior details once buoyancy is large.
► Heat transfer deterioration may be qualitatively predicted only by complex models.
► Isotropic and anisotropic turbulence models perform alike in supercritical flows.