Abstract
Internal corrosion during the oil transportation process has been long associated with the interaction of free water with the pipe surface. As a result, prediction of water phasic distribution along the flow line, especially the information whether oil or water wets the pipe wall is crucial in order to identify mitigation strategies and establish overall pipeline integrity. Use of computational techniques can not only address the wettability issues but also help identify mitigation criteria and its effective deployment.
In the present study, detailed Computational Fluid Dynamics (CFD) modeling of turbulent oil-water multiphase flow, with water cut of 20% or lower, is carried out in a horizontal carbon steel pipeline to predict different water wetting regimes as well as corrosion kinetics. For this purpose, Eulerian multiphase models with appropriate sub-models for interfacial forces, turbulent interaction and population balance equations have been exercised to address the dispersed (water) phase dynamics such as droplet breakage and coalescence in a detailed fashion. In addition, generation of ferrous ions (Fe2+), due to metal-water interactions, is modeled by including detailed electrochemistry and corrosion kinetics.
The numerics captured various water wetting regimes inclusive of intermittent, fully wetting, dispersed and semi-dispersed water-in-oil behavior. Results were in very good agreement with the experimental recordings and the calculations clearly predicted that for given superficial water velocity, elevating the superficial oil velocity leads to transition of regimes from stable water wetting to intermittent water wetting and subsequently stable oil wetting scenario. Phasic fraction and velocity distribution at different flow cross-sections reveal the flow modulations as a function of the turbulence inherent to the two-phase system. Flow coupled corrosion calculations clearly indicate a substantial generation of Fe2+ ions under stable water wetting regime, while the ferrous ion production levels deteriorate with intermittent water wetting, an effect well characterized in flow loop tests.
