Modelling of aqueous carbon dioxide corrosion in turbulent pipe flow

• Main Author: Wang, Fangyou
• Other Authors: Postlethwaite, John
• Format: Thesis
• Language: English
• Published: University of Saskatchewan 1999
• Subjects: Carbonic acid
• Online Access: http://hdl.handle.net/10388/etd-10212004-002022

A predictive model is described to simulate the aqueous CO2 corrosion of iron in turbulent pipe flow. Mass transfer is modelled with a two dimensional low Reynolds number k- 3 turbulence model by simultaneously solving the conservation equations for mass, momentum, kinetic energy of turbulence and its dissipation rate, along with the concentrations of various dissolved species. The effect of the slow homogeneous chemical reaction Of CO2 hydration is incorporated into the model by including an extra source term in the transport equation for carbonic acid. Other homogeneous chemical reactions are assumed to be in equilibrium. The electrochemical reactions considered are iron dissolution and the reduction of H2CO3, H+ and H 2O. An iterative procedure is employed to calculate CO2 corrosion rates by ensuring the mixed potential theory is satisfied and the various surface chemical equilibria are preserved. It is shown that at a bulk PH 4 and very low flow velocities the reduction of H2CO3 is controlled by CO2 hydration. At higher velocities and in the high current density region where Tafel behavior is no longer observed, H2CO3 reduction is controlled by the interaction of chemical reaction (CO2 hydration) and mass transfer. The H+ reduction is more flow sensitive than the H2CO3 reduction. CO2 corrosion is more influenced by flow at PH 4 than at higher PH values. Generation of H 2CO3 by CO2 hydration enhances the H2CO 3 mass transport and reduces the H2CO3 mass transfer entrance length. The relative contributions of H2CO3, H+, find H2O reduction to the corrosion current depend on the solution PH. The CO2 corrosion rate is accelerated by turbulent flow and CO2 partial pressure through an approximate power law. Present model predictions were tested against experimental findings and other models. Good agreement is achieved. A simplified spreadsheet-based electrochemical model of CO2 corrosion is also developed. Overall mass transfer coefficients from established correlations are used. From the user input data (T, pH, Ub, PCO2 , d, etc.), corrosion rates are calculated based on the mixed potential theory. Macros and controls are used to in the model to automate the calculation process.