Kevin L. Heppner

Effect of the Crevice Gap on the Initiation of Crevice Corrosion in Passive Metals

Abstract Past research into the mechanism governing the time to active crevice corrosion—the incubation period—of a passive metal crevice has produced theoretical models coupled with the B-dot model, the Debye-Huckel limiting law, and other activity models to correct for nonideal behavior at moderately high concentrations. In this research, the transport model of Watson and Postlethwaite is coupled with the ionic interaction model of Pitzer to predict the effect of the crevice gap on the iR drop and chemical activity of the crevice solution. Two cathodic reactions, crevice external oxygen reduction and crevice internal hydrogen ion reduction, are assumed to balance metal dissolution. To validate the model, the experimental Type 304 (UNS S30400) stainless steel crevice of Alavi and Cottis is simulated. Model predictions improve upon predictions of past models and match observations of this experimental work within experimental uncertainty. The effect of crevice gap on a titanium crevice immersed in 0.5 M a...

Effect of Ionic Interactions on the Initiation of Crevice Corrosion in Passive Metals

In the crevice corrosion process, oxygen reduction occurs faster than diffusion into an occluded crevice, causing deoxygenation of the crevice solution. Once oxygen is depleted in the crevice, oxygen reduction can only occur on the metal surface outside of the crevice. The cations produced by metal dissolution are hydrolyzed, which causes the pH of the crevice solution to drop. This increases the rate of metal dissolution, forming an autocatalytic coupling that causes concentrated electrolyte solutions to form in the crevice. The focus of this paper is the prediction of the effect of nonideal solution behavior on crevice corrosion using the ionic interaction model of Pitzer coupled with an electrolyte mass-transport model. This mathematical model was used to simulate the type 304 stainless steel crevice corrosion experiment of Alavi and Cottis. The results are in excellent agreement with experimental observations. The model was then applied to simulate the crevice corrosion initiation period of a titanium crevice. Comparison of the predictions to those generated via an ideal solution crevice corrosion model indicates that interionic forces draw chloride ions into the crevice and hinder the transport of hydrogen ions out of the crevice.

Flow-Accelerated Corrosion Susceptibility Prediction of Recirculating Steam Generator Internals

Steam generator (SG) components are subjected to corrosive solutions in turbulent flow. Under such conditions, actual component lifetimes may be significantly reduced from their original design lifetimes. Premature replacement of steam generator components before their expected lifetime can be very expensive. Furthermore, degradation of essential components can reduce the steam generator efficiency, thus reducing net profits. Moreover, a SG failure can also be a safety issue. One of the degradation mechanisms affecting secondary-side SG internal structural elements, which are referred to as internals, is Flow-Accelerated Corrosion (FAC). The susceptibility to FAC depends on flow parameters, water chemistry, and materials. All SG internals made of carbon steel are susceptible to FAC to varying degrees. For FAC susceptibility prediction, flow velocity, pH, and oxygen distributions are needed. SG codes, including THIRST (T hermal H ydraulic analysis I n ST eam generators, a computer code developed by AECL), traditionally solve for thermalhydraulic parameters. A new chemistry module has been added to THIRST, which now makes this code useful for the prediction of local water chemistry parameters in the SG. The THIRST chemistry module is comprised of a multicomponent, multiphase mass transport model coupled with a multiphase chemical equilibrium model. As input, the module requires amine concentrations in the feedwater and reheater drains. The module predicts local distributions of amine concentration in the secondary side. The concentrations predicted by the module are used to compute the pH. The chemistry module was verified against results of other work in the literature and against station blowdown data. Flow and chemistry predictions of THIRST were used to predict FAC susceptibility for internals of a SG with an integral preheater and a SG without it. Ranking of SG locations in order of FAC susceptibility was estimated from an empirical, Kastner-Riedle model. The most susceptible internals are predicted to be those in the upper section of the hot side and those on the cold side that are near the SG centre, while SG lower regions, including the integral preheater, if one exists, are better protected.Copyright

New method for calculating charge density in electrochemical systems

Abstract A new model for calculating the influence of charge density on electrolyte mass transport has been developed. This model is derived from Poissons equation for charge density and is implemented as an algebraic charge density correction. Use of this method greatly improves the computational efficiency of electrolyte mass transport modelling by avoiding the solution of Poissons equation. The new correction was used in the simulation of a moving boundary experiment. The predicted moving boundary velocity matched experimental results. Furthermore, the performance of this model was compared with the operator splitting algorithm of Evitts and Watson. Use of the new charge density model resulted in improved computational efficiency, numerical stability and accuracy.

A hybrid differencing scheme for mass transport in electrochemical systems

Purpose – To present a new hybrid differencing scheme for the numerical solution of an electromigration‐diffusion equation. The value of this work is evidenced by demonstrated improvement in the simulation of the Fu and Chan experiment when using the hybrid scheme.Design/methodology/approach – A hybrid differencing scheme is developed which is based upon the solution of the pseudo‐steady state electromigration‐diffusion equation. In this scheme, a weighting parameter is calculated that varies the relative influence of the upwind node (relative to the direction of electromigration). This scheme significantly enhances the accuracy of electrochemical system mass transport models.Findings – The hybrid scheme was compared to the upwind scheme. Use of the new hybrid scheme improved the accuracy of the model predictions by as much as 87 percent compared to the upwind scheme. However, use of the new scheme also increased the simulation time by between 6 and 43 percent. Deviations from electroneutrality and the pr...

Modelling of the effect of hydrogen ion reduction on the crevice corrosion of titanium

In this research, the effect of the level of hydrogen ion reduction on a titanium crevice immersed in 0.5 M NaCl solution at 70°C was investigated. The fraction of the total dissolution current supplied by hydrogen ion reduction, Ψ, was varied from 0 to 0.8 in increments of 0.2 and the steady state pH, conductivity (κ), and iR drop (φ) profiles in the crevice were calculated. The mass transport model of Watson and Postlethwaite was employed and the crevice was assumed to be passivated. As Ψ increased, the solution conductivity and the iR drop along the crevice were significantly reduced while the pH showed only a slight increase. A plot of total iR drop along the crevice length, pH at the crevice tip, and solution conductivity at the crevice against Ψ was constructed and the following relationships were obtained (R 2 >0.995): pH=0.0091ψ+1.27 p H = 0.0091 ψ + 1.27 Φ=−0.919ψ+97.0 Φ = − 0.919 ψ + 97.0 κ=−0.0044 ψ 2 −0.399ψ+157 κ = − 0.0044 ψ 2 − 0.399 ψ + 157 Of the three solution properties studied, the pH showed the weakest dependence upon the hydrogen ion reduction fraction, varying linearly from 1.27 to 2.01 as Ψ varied from 0% to 80%. The iR drop along the crevice length showed the greatest sensitivity to the hydrogen ion reduction level.