Nucleation of super-critical carbon dioxide in a venturi nozzle Academic Article uri icon

abstract

  • © 2016 Elsevier B.V. Pressure reduction at the entrance of the compressor in supercritical CO2 Brayton cycles may cause nucleation and create a mixture of vapor and liquid droplets due to operation near the saturation conditions. Transient behavior of the flow after nucleation may cause serious issues in operation of the cycle and degrade the materials used in the design. The nucleation behavior of supercritical carbon-dioxide inside a venturi nozzle near the critical point is computationally studied. A transient compressible 3D Navier–Stokes solver, coupled with continuity, and energy equations have been implemented. In order to expedite the simulations, Fluid property Interpolation Tables (FIT) based on a piecewise biquintic spline interpolation of Helmholtz energy have been integrated with OpenFOAM to model S-CO2 properties. The mass fraction of vapor created in the venturi nozzle has been calculated using homogeneous equilibrium model (HEM). Nucleation behavior has been shown to be very sensitive to the inlet pressure, inlet temperature, and flow rate. The flow conditions that led to nucleation were identified. Nucleation was observed in the throat area and divergent section of the nozzle for mass flow rates from 0.050 kg/s to 0.065 kg/s, inlet pressure from 7.8 to 7.4 MPa for fixed exit pressure equal to 7.28 MPa. The inception of high-vapor-content nucleation was first observed in the throat area away from the side walls that remained confined to the throat region in later times. However, near the walls, a high liquid-content two-phase region was detected, first in the divergent section. At later times, the two-phase region was convected downstream toward the nozzle exit. Nucleation in central plane was associated with larger pressure drop and higher vapor-content (higher volume fraction); whereas lower pressure drop and more liquid-content (lower volume fraction) was observed near the walls.

author list (cited authors)

  • Jarrahbashi, D., Pidaparti, S. R., & Ranjan, D.

citation count

  • 3

publication date

  • December 2016