Question 5 of 6: Stress Assisted Corrosion

Question 5: Stress Assisted Corrosion (20 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

5.1 — (a) Stress corrosion cracking: conditions, mechanism, and engineering solutions

Conditions. Stress corrosion cracking (SCC) needs three factors present at once, and removing any one stops it: a susceptible alloy in a specific environment (SCC is famously alloy/environment-specific rather than general — austenitic stainless steels crack in hot chloride solutions but are essentially immune to SCC in ammonia, while brasses crack in ammoniacal environments but not typically in chlorides); a sustained tensile stress, at or below the yield strength, which may be entirely from residual stress (welding, cold forming, grinding) with no external service load at all; and enough time at the combined condition, often with a significant, hard-to-predict incubation period before visible cracking begins. Typical alloy/environment pairs relevant to this exam code: austenitic stainless steels in hot (above roughly 60 °C), chloride-containing water; brasses in ammonia or amine environments ("season cracking"); high-strength steels in H₂S or other hydrogen-bearing environments (hydrogen-induced SCC); and aluminum alloys in humid air or chloride solutions along specific short-transverse grain-boundary paths.

Mechanism. Two mechanisms operate, sometimes together: (1) anodic-dissolution-controlled SCC, in which the tensile stress ruptures the protective passive film at a highly localized, crystallographically favoured site (often a slip step where an emerging dislocation breaks the film), exposing bare, highly anodic metal beneath; the exposed metal repassivates, but the cycle of film rupture → local dissolution → repassivation repeats preferentially at the crack tip (which is continuously strained by the applied stress) far faster than anywhere else on the surface, so the crack advances by a ratchet of micro-dissolution events rather than uniform corrosion, typically along an intergranular or a specific transgranular (crystallographic) path; and (2) hydrogen embrittlement, in which atomic hydrogen produced by the corrosion reaction (or by cathodic protection over-current, or absorbed from a sour/H₂S environment) diffuses into the metal ahead of the crack tip, concentrates at the region of maximum triaxial stress just ahead of the tip, and locally embrittles the lattice (by mechanisms including hydrogen-enhanced decohesion and hydrogen-enhanced localized plasticity), so the crack advances by brittle micro-fracture rather than by dissolution at all. In both mechanisms the crack propagates approximately normal to the applied tensile stress, at stresses well below the material's normal fracture strength, and with essentially no ductile deformation or visible general corrosion elsewhere on the part — which is exactly what makes SCC dangerous: a component can fail suddenly with almost no external warning.

Engineering solutions. (i) Change the alloy to one known to be immune or highly resistant in the specific service environment (e.g. duplex or higher-nickel stainless in place of austenitic 300-series for hot chloride service); (ii) reduce or eliminate the sustained tensile stress — stress-relief anneal welds and cold-formed parts, avoid tight-fit/press-fit assembly stresses, and design to keep the maximum service stress below the alloy's known SCC threshold stress, $\sigma_{SCC}$; (iii) introduce a compressive residual surface stress (shot peening, surface rolling), since SCC is a tensile-stress-driven mechanism and a compressive surface layer must first be overcome before a net tensile condition exists at the surface; (iv) modify the environment — lower temperature, remove or dilute the specific aggressive species (chloride, ammonia, sulfide), control pH, add inhibitors, or exclude oxygen where the mechanism is oxygen-dependent; and (v) apply cathodic protection carefully — it suppresses anodic-dissolution-controlled SCC but, if over-applied, can promote hydrogen-embrittlement-controlled SCC on susceptible high-strength alloys, so the protection potential must be selected with the dominant mechanism in mind.

5.2 — (b) Corrosion fatigue: conditions, mechanism, and engineering solutions

Conditions. Corrosion fatigue needs a cyclic (fluctuating) stress, with a tensile component, applied to a metal in a corrosive environment — but unlike SCC, essentially no metal/environment combination is immune: corrosion fatigue is a general phenomenon that can occur in almost any alloy exposed to almost any corrosive medium (fresh or salt water, humid air, process chemicals), not a specific-pairing effect. It differs from SCC in a second important way: no threshold stress is required below which failure never occurs.

Mechanism. Ordinary fatigue in dry air already nucleates cracks at persistent slip bands or surface stress concentrations and grows them at a rate governed by the Paris law $da/dN=A(\Delta K)^n$. A corrosive environment accelerates both stages: at initiation, the same repeated slip-band film-rupture/repassivation cycle described for SCC above roughens the surface and nucleates cracks far earlier (and at far lower cyclic stress amplitude) than in dry air, removing the fatigue limit that ferrous alloys otherwise show — in a corrosive environment the S–N curve for a normally fatigue-limited steel instead keeps sloping downward indefinitely, so there is no stress amplitude, however small, that guarantees infinite life. At propagation, each tensile excursion of the cycle re-exposes bare metal at the crack tip (mechanically rupturing whatever oxide/passive film had reformed during the previous cycle) to fresh corrosive attack, and, for susceptible alloys, hydrogen generated by the corrosion reaction can additionally embrittle the crack-tip region exactly as in hydrogen-assisted SCC; the combined effect raises the crack growth rate at a given $\Delta K$ well above the inert-environment Paris-law rate, and can also lower the apparent threshold $\Delta K_{th}$. The net effect is a fatigue life that can be an order of magnitude or more shorter than the same loading history in air.

Engineering solutions. (i) Reduce the cyclic stress amplitude and eliminate stress concentrations (fillet radii, smooth surface finish, avoid sharp re-entrant features) so far fewer sites are available to nucleate a corrosion-fatigue crack; (ii) apply a compressive residual surface stress (shot peening, surface rolling, induction hardening) for the same reason given under SCC — a compressive layer both raises the effective threshold and slows early crack growth through the near-surface region; (iii) modify the environment (inhibitors, deaeration, coatings that exclude the corrosive medium from the metal surface, corrosion-resistant cladding); (iv) apply cathodic protection, which is generally more effective and less prone to unwanted side effects against corrosion fatigue than against SCC, since corrosion fatigue does not have the same hydrogen-embrittlement sensitivity concern for most structural steels; and (v) select a more corrosion-resistant alloy or apply a protective coating/plating specifically over the fatigue-critical region, combined with routine inspection intervals set from a fracture-mechanics life calculation that uses the accelerated (environment-specific) $da/dN$ data rather than the in-air Paris-law constants.