Book Description

A numerical simulation campaign is conducted to better elucidate flow physics and modeling requirements of a supercritical (SC) nitrogen jet injected into a tank of quiescent sC nitrogen. The goals of this work are twofold: to inform the design of injectors and power combustion chambers for use in the direct-fired supercritical CO2 (s-CO2) generation cycle and cryogenic liquid propellant rockets, and to investigate the extent to which meaningful flow characterization can be achieved with computationally expedient methods, using commercial software. Reynolds-Averaged Navier-Stokes (RANS) and Large Eddy Simulation (LES) approaches are used in STAR-CCM+ versions 10.06.010 and 12.02.011. Jet disintegration is evaluated with velocity, density and temperature profiles, potential core penetration and identification of turbulent length scales. These data are compared with experimental data and evaluated against other modeling approaches. Mixing behavior is expected to mimic that of a single-phase jet, and be diffusion-driven, as there will be no droplet formation in the supercritical phase. Challenges are encountered in high computational requirements inherent to unsteady LES. Challenges are also encountered in simulation stability and convergence given large flow gradients near jet exit, large fluid property gradients near the critical point, and the small length scale of energetic flow features unique to this high-pressure thermodynamic regime. Simulation results over-predict core penetration compared to experiment and previous numerical efforts and show an overall slower transition to ambient conditions. It is shown however that commercial code can correctly synthesize the overall flow physics and trends of the single-phase gas jet behavior expected in purely supercritical turbulent mixing flow.