Technical Presentations at the October 2011 Meeting

1.1    ‘Evolution of Cracks from Corrosion Pits’, Alan Turnbull, NPL 

In many engineering applications, pitting is the precursor to stress corrosion cracking (SCC) as it provides the required combination of an aggressive local solution chemistry and a stress concentrating feature (the conventional perspective). The fundamental steps in the overall process of crack development involve pit initiation, pit growth, the transition from a pit to a crack, short crack growth and long crack growth. A reliable prediction of the complete damage process in engineering components would require a quantitative understanding of each stage.  A number of deterministic (mechanistic-based) models has been developed to address SCC initiation and propagation form pits. In these predictive schemes, the pit-to-crack transition is based on the phenomenological requirements proposed by Kondo for corrosion fatigue: that the pit depth must be greater than a threshold depth and that the crack growth rate should exceed the pit growth rate. Implicit also is the assumption that the pit is actively growing. In simple terms, a stable crack will emerge from the pit if the threshold stress requirement for crack development is exceeded (the pit acting as a concentrator of stress and strain) and if locally the crack growth is sufficient for the crack to outrun the dissolving pit surface at that location. At the point of transition from a pit to a crack, the crack is considered effectively to have the same depth as the source pit. 

The validity of this latter proposition has been examined in detail for a steam turbine disc steel exposed to sustained upset conditions of aerated 1.5 ppm chloride (under normal operating conditions the steam turbine is aerated only off-load and transiently as the system comes on-load; on-load the condensate is oxygen free and the chloride concentration is closer to 300 ppb). Using the Skyscan X-ray tomography instrument at U. Birmingham, remarkable 3‑D microtomographic images have been obtained that demonstrate that cracks actually develop predominantly at the shoulder of the pit, at or below the pit-surface interface, for specimens stressed to 50%, 70% and 90% of s0.2.  Thus the assumption of a crack depth equivalent to the pit depth at the point of transition, implying the crack initiates at the pit base, has no foundation (at least for this system).  

An explanation for initiation of SCC near the pit mouth based on electrochemical differences between and pit mouth and base was not tenable and a mechanics-based explanation was sought.  To explore this, finite element (FE) analysis was undertaken of model pits with hemispherical and “bullet-shaped” geometries in a cylindrical rod subjected to longitudinal applied stresses ranging from 10%-90% s0.2. At the high stresses relevant to service, the principal strain was predicted to be a maximum in the material just below the pit mouth and included a significant plastic component (reduced constraint to plastic deformation near the surface). The principal stress was lowest in this region. The localisation of plastic strain to just below the mouth would in itself be argued as a basis for the location of stress corrosion crack initiation. However, in another novel feature of this work, it was recognised that if the pit is growing in a static stress field that generates plastic strain then that plastic strain is dynamic.  Since slow strain rate testing is recognised as being the most sever test condition for evaluation of stress corrosion cracking, here we have a situation in which the material local to the pit is undergoing slow plastic strain due to the pit growth. FE simulation of a growing pit in a static stress field indicated corresponding plastic strain rates that were commensurate with values associated with stress corrosion cracking. This observation introduces a wholly new concept in understanding the evolution of stress corrosion cracks from pits. 

Thus, it may be suggested that the role of growing pits in initiating SCC is not simply to produce a local aggressive environment and a localisation stress and strain but most critically to induced dynamic straining of adjacent material.  [Acknowledgements: L. Wright, L. Crocker and S.Zhou: NPL; A.D. Horner, B. Connolly: U. Birmingham]

  1.2  Coating of Thermally Sprayed Metal Substrates’, Malcolm Morris, Leighs Paints

The presentation gave an outline of thermal metal spray as a long established method of corrosion control, including common types and methods of application.  The morphology of the thermal sprayed film was discussed; and due to the porous nature of the film, it is common practice to apply a coat of paint in order to seal the porosity. 

The sealer coat is not decorative and does not have an overlay on the surface of the metal spray, therefore the concept of ‘duplex’ systems of metal spray overcoated with high build coating specifications has been common practice, both in order to enhance the longevity of the corrosion protection, and also to give better aesthetic appearance. 

Duplex systems of thermal srayed zinc (TSZ) have a long track record, however duplex thermal sprayed aluminium systems have shown examples of failure in service, with premature detachment of high build coatings from the TA, and exhaustion of the aluminium coating. 

A mechanism for the failure of these duplex systems under high chloride concentrations was discussed, whereby aluminium chloride is converted to hydrochloric acid and the electrolytes are trapped within the TSA by the high build paint system, leading to rapid degradation and detachment of the coatings.

 

Back to Minutes

  4.1  'New Investigations on Wall Shear Stresses of Copper Nickel Alloys in Natural and Artificial Seawater' Christos Kapsalis, KME Germany AG & Co. KG. 

Recent investigations on critical wall shear stresses for CuNi 90/10 in artificial ASTM seawater revealed that CuNi 90/10 can endure much higher wall shear stresses than is generally reported in the literature.  These results raised again the question on the validity of corrosion results obtained in ASTM seawater for the prediction of materials performance in natural seawater.  The major difference is the formation of biofilms known to play a significant role.  The justifications for the previous experiments in ASTM seawater were that the flow velocities applied were above the adherence limits of marine biofilms and that the focus of the work was on the effect of disinfectants, in the presence of which biofilms should not exist.  Nevertheless, a comparative study was launched to evaluate the critical wall shear stresses of CuNi 90/10 and CuNi 70/30 in untreated and chlorinated artificial ASTM seawater and natural sea water.   

The results confirmed that the CuNi alloys perform very similar in both environments and that the flow resistance of these alloys is significantly higher that reported in the literature.  The resistance to flow induced localized corrosion (FILC) depends on the alloy composition, the nature of protective corrosion product layers formed during different surface treatment procedures and the presence and concentration of hypochlorite.

 

 4.2  ‘Pitting Corrosion of Stainless Steel: Measuring and Modelling Pit Propagation’, Majid Ghahari, University of Birmingham School of Metallurgy and Materials 

Stainless steel containers are used for interim storage and planned for final disposal of intermediate level radioactive waste in a geological disposal facility.  Although stainless steel is generally resistant to corrosion, there is concern that, during interim storage above ground and in the operational phase of the facility, pitting corrosion due to chloride salts deposited from aerosols may take place.  It is therefore important to develop and validate models for the prediction of corrosion damage over relatively long timescales. In order to predict the rate of pit growth, it is necessary to understand the mechanism of pitting corrosion, develop prediction models and validate them by comparison with experiment.  

Highly intense X-rays from a synchrotron are an ideal probe for in situ investigation of the evolution of corrosion processes, since they are able to penetrate water and even metal.  The corrosion behaviour of 2D pits growing in the edge of stainless steel foil has been observed in situ using fast radiographic imaging, showing low growth at the pit bottom, but higher rates of growth in the lateral direction, undercutting the surface and producing a characteristic “lacy” pit cover.  A similar morphology has been observed in 3D by X-ray microtomography.   

The local corrosion current density in the 2D pits can be determined from the rate of advance of the pit into the metal. The local solution composition can be backcalculated from the ions injected into the solution by dissolution and their transport out of the pit.  The local potential distribution can also be determined.  Thus the local current density as a function of local metal ion concentration and interfacial potential along pit boundary has been calculated for 304 SS, and compared with the values previously published for 1D artificial pits.  

It is shown that the shape of pits depends on the solution chemistry: in dilute chloride concentrations, the pit cover provides a diffusion barrier and maintains sufficiently aggressive chemistry inside the pit cavity for stable growth. It is also shown that pits do not necessarily maintain their hemispherical/dish-shape during growth and more tunnel-like morphologies may develop under  current-limited conditions. In these cases, the actively corroding area of the pit does not increase with depth, suggesting that pits can regulate their growth to fit the supplied current. A cathodic current limitation, therefore, would not necessarily support an upper bound on the pit depth, although the electrochemical potential drop owing to current transport in deeper pits may do so.  

The results from the in situ real time synchrotron experiments have been used to refine an existing model of pit propagation in stainless steel.  [msg713@bham.ac.uk]

 Back to Minutes