The following information may be useful for your research

Abstract: Gas assisted atomization is becoming increasingly important in many industrial applications such as physical,chemical and petroleum processes. In order to achieve proper atomization it is crucial to have proper mixing of gas (air)and liquid (water) in the feeding conduit before it enters into the nozzle. The flow regime, as well as the flow pattern andstructure of the flow, are some of the important parameters that describe two-phase gas/liquid flows, and identify twophase

gas/liquid flow regimes. It is also desirable to know under what conditions there is a transition among the differentflow regimes (dispersed, stratified, annular, annular-dispersed, slug, wavy-slug, mist-annular). Due to the existence ofrelative movement in the interfaces and variable interactions between two phases, two-phase gas/liquid flow is a complextransport phenomenon compared to single-phase flow. Still, there is no effective technique to identify the two-phasegas/liquid flow regimes and it is even difficult to capture the accurate flow structures in smaller conduits in turbulent flowcases. Lack of solid and comprehensive theories for predicting and calculating the pressure and void fraction variations intwo-phase air/water flow situations has left engineers without essential information for proper design of two-phase flowsystems. This review is an effort to explore the state of the present advanced measurement techniques in this field ofresearch. Subsequently, some of the advanced void fraction, photonics and pressure measurement techniques and

correlations for identification of two-phase gas/liquid flow regimes and bubble sizes are investigated

RefA Critical Review of Advanced Experimental Techniques to Measure TwoPhase

Gas/Liquid Flow

M.A. Rahman, T. Heidrick and B.A. Flec

The Open Fuels & Energy Science Journal, 2009, 2, 54-70

benthamopen.com/toefj/articles/V002/54TOEFJ.pdf

I would first consult some of the classic Chem Eng reference and textbooks, e.g. Bird/Stewart/Lightfoot, McCabe, Perry or Coulson/Richardson. Two-phase flows and their effect on mass transfer etc. has been exhaustively studied and taught (not to say that more understanding isn't needed!).

Thanks for your contribution. The article is interesting it is a good summery of the state of the art, but I did not find hints regarding my question. Most of the models are based on void fraction and quality measurements/predictions. Nobody seems to care about the interface area between gas and liquid. The conclusion is, this is not necessary.

But think about interaction of the phases. Shear stress is located on the phase interface and shear stress times surface area results in a shear force. Less area between phases results in  less force between them even if the shear stress is the same.

I know that in practice it is very difficult  to measure the surface area but void fraction is not easy too and there are measurement methods. In a lot of processes where mass transfer between phases is involved there are measurements or estimations of surface area density but I wonder whether researchers tried to model flow pattern transition or simple pipe flow with regard to surface area.

As you can see, my thoughts and ideas still spinning around.

@jan: Yes may be you are right. I still do not know all about two phase flow and may be I am missing a convincing idea that makes my thoughts obsolete. But honestly said: I am getting bored with all that old stuff. We all know these dimensionless numbers Fr, We, Re, w* etc... but do you have the feeling that it really explains what happens within two phase flow?? Ok, there are well understood flow patterns like bubbly flow or annular flow. But in my case I often have to think about new geometries like criss cross structure inside plate heat exchangers.  In cases where the geometry is new you are mostly lost. The first thought is: Lets build a flow pattern map and see what the phases are up to. This is why I am searching for new studies that focus on a deeper understanding of the phase interaction process. Why not trying to predict where and how gas and liquid are flowing, there must be physical reasons.

But this has not much to do with the question I have asked above, sorry.

In case of the reference paper, there is a liquid flow gravitationally driven in counter current with a gas flow. If the gas static pressure gradient is comparable with liquid density x gravitational acceleration x cos(pressure gradient vector to vertical) then one may have an unstable condition, i.e. the liquid tending to slow down. SInce the local accumulation of liquid increases the gas static pressure gradient, the flow is inherently unstable. The details of wetting angles and surface configuration may drive the local stability (so superficial tension do play a role). I don't think that there is a general answer to your question.