Quantification and Prediction Tools for Atmospheric Corrosion and Electrochemistry in Non-Conventional (Micro-)Setups

Atmospheric corrosion predictions for design and engineering require time-consuming exposure experiments and provide only a rough estimation of the corrosion rate. At the other hand, the multi-scale problem of atmospheric corrosion involves thin film electrolytes, evaporating -highly concentrated- solutions where local variations are controlling the overall process. Conventional (bulk) electrochemistry is therefore inadequate for the quantification or even the understanding of atmospheric corrosion mechanisms. The Multi-Ion Transport and Reaction Model (MITReM) is the link between the underlying electrochemical physics and corrosion rate predictions. With a numerical finite element method we simulate the effect of electrolyte thickness, electrode geometry and NaCl concentrations during evaporation in ultra-thin (sub 50 µm) electrolytes. The zinc corrosion rate in these thin films is function of the calculated local, time-dependent variations in the electrolyte. The two-dimensional simulations demonstrate the geometrical effect on the local current density and its deviation from simple limiting current equations. The time-dependency also describes the evolution of the electrolyte concentrations and the current transients. The simulated formation of corrosion products is compared with experimental measurements of the initial oxidation of zinc in humidified air. The evaporation/condensation dynamics of ultra-thin films in combination with complex electrolyte compositions introduces high computational costs. The application of these models is therefore limited to small or highly simplified geometries. It also implies limits on the time scales that can be simulated. To reduce the computational cost a decoupling method is developed to decrease the number of finite elements and hence the number of unknowns in the system. The transport in the direction of the electrode plane and the perpendicular direction are decoupled. The decrease in computational cost enables numerical simulations of complex electrochemical systems on extended complex geometrical shapes and for longer time scales. The decoupling method has an additional benefit; the complexity of moving triangular meshes is avoided since the electrolyte height geometry is a combination of one-dimensional meshes. The reduced decoupling model is compared with the full-scale simulations of a galvanic couple (zinc-steel). The results show that the decoupled model approaches the full-scale model for thinner electrolyte layers. This observation is supported by the evolution of the edge-effect, illustrated with two-dimensional simulations. In thin layers, the minimal accuracy loss is acceptable in comparison to the gain in computational time. The use of the dimension reduction for the simulation of atmospheric corrosion conditions is therefore justified. The presented work improves the developments of quantification and prediction tools for atmospheric corrosion and electrochemistry in non-conventional (micro-)setups. A better description and prediction of the corrosion process is obtained in ‘dynamic’ thin electrolyte layers.