Closure relations effects on the prediction of the stratified two-phase flow stability via the two-fluid model

R. Kushnir, V. Segal, A. Ullmann, N. Brauner*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


The possibility of predicting the exact long wave linear stability boundary via the two-fluid (TF) model for horizontal and inclined stratified two-phase flow is examined. The application of the TF model requires the introduction of empirical closure relations for the velocity profile shape factors and for the wave induced wall and interfacial shear stresses. The latter are recognized as the problematic closure laws. In order to explore the closure relations effects and to suggest the necessary modifications that can improve the stability predictions of the TF model, the results are compared with the exact long wave solution of the Orr–Sommerfeld equations for the two-plate geometry. It is demonstrated that with the shape factors corrections and the inclusion of wave induced stresses effects, the TF model is able to fully reproduce the exact long wave neutral stability curves. The wave induced shear stresses in phase with the wave slope, which give rise to the so called “sheltering force”, were found to have a remarkable destabilizing effect in many cases of horizontal and inclined flows. In such cases, the sheltering effects must be included in the TF model, otherwise the region of smooth stratified flow would be significantly over predicted. Based on the results of the exact analysis, a simple closure relation for the sheltering term in the TF model is provided.

Original languageEnglish
Pages (from-to)78-93
Number of pages16
JournalInternational Journal of Multiphase Flow
StatePublished - Dec 2017


  • Closure relations
  • Multiple solutions
  • Sheltering
  • Stability
  • Stratified flow
  • Two-fluid model


Dive into the research topics of 'Closure relations effects on the prediction of the stratified two-phase flow stability via the two-fluid model'. Together they form a unique fingerprint.

Cite this