Calculation of mass transfer coefficients for corrosion prediction in two-phase gas-liquid pipe flow

•Methodology for estimation of mass transfer rates in gas-liquid flow is introduced•Chilton-Colburn analogy and near wall eddy diffusivity distribution are considered•Mechanistic gas-liquid flow modeling is used to obtain liquid flow characteristics•The models agree well with experimental data from large-scale gas-liquid flow Internal corrosion in industrial environments involving gas-liquid flow can be a serious concern. For example, in the oil and gas industry, corrosive water phase is usually transported in pipes along with liquid and/or gas hydrocarbon phases. These gas-liquid flows develop complex flow patterns where the liquid phase can distribute in quite different ways (as stratified layers, intermittent slugs, annular film, etc.) depending on the gas and liquid flow rates and pipe inclination. In these circumstances, the water phase can flow at very high velocities leading to high turbulence and mass transfer rates that can accelerate corrosion of the metallic pipe surface. In general, proper prediction of corrosion rates via mechanistic electrochemical models requires the knowledge of the mass transfer rate of corrosive species in the aqueous phase. There are very few studies in the open literature that show specific experimental data or propose ways to compute mass transfer rates in gas-liquid flow; particularly, for large pipe diameters. The present study introduces a methodology for the estimation of mass transfer rates in gas-liquid flow via the Chilton-Colburn analogy and near-wall eddy diffusivity distribution based on the mechanistic gas-liquid flow modeling. The proposed mechanistic models cover a wide range of fluid's properties and flow rates, different flow patterns and pipe inclinations, and show good agreement with mass transfer experimental data from large-scale gas-liquid flow.