Question 3 of 6: Metallurgical Cells

Question 3: Metallurgical Cells (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.

3.1 — (a) Galvanic corrosion: conditions, mechanism, and engineering solutions

Conditions. Galvanic corrosion needs three things present simultaneously: two electrochemically dissimilar conductors (usually two different metals, but a single alloy with two distinct microstructural phases, or the same metal in two different states of surface finish or aeration, can also form a galvanic couple); an electrolyte that wets both and is continuous between them, so that ionic current can flow; and a direct electrical (metallic) connection between the two conductors, so that electron current can return. Remove any one of the three — insulate the joint, keep the assembly dry, or use metals close together on the galvanic series — and the mechanism cannot operate. The severity of the attack is set by three further factors: the potential difference between the two metals in the actual service electrolyte (the galvanic series in seawater, not the standard electrochemical series in distilled water, is the relevant ranking, because alloy composition, surface films and the specific electrolyte all shift real-world potentials); the relative area ratio of cathode to anode (a small anode coupled to a large cathode concentrates all the cathodic current onto a small anodic area, multiplying the local corrosion rate — steel rivets in a large copper plate corrode far faster than the reverse pairing); and the electrolyte's conductivity and the distance between the two metals (attack is worst immediately at the junction and falls off with distance in a low-conductivity electrolyte, but can be more evenly spread in a highly conductive one such as seawater).

Mechanism. The more active (less noble) metal becomes the anode of the couple and its own corrosion current is augmented by the current demanded by the cathodic reaction occurring on the more noble metal's much larger, more efficient surface; the noble metal is cathodically protected (its own corrosion all but stops) while the active metal corrodes faster than it would alone. This is precisely the mixed-potential picture of Question 2: coupling shifts the whole system to a single compromise (mixed) potential between the two metals' individual free-corrosion potentials, at which the total anodic current from the active metal must equal the total cathodic current supplied (mostly) by the noble metal's surface.

Engineering solutions. (i) Select metals close together in the galvanic series for the specific service electrolyte, or use compatible metal combinations recommended by design charts; (ii) electrically insulate the joint (dielectric unions, non-conductive gaskets/washers/sleeves at bolted or flanged connections) to break the metallic path; (iii) avoid an unfavourable area ratio — if dissimilar metals must be joined, make the anodic (active) member the larger area and the cathodic member as small as possible (e.g. use coated fasteners of the noble metal in a structure of the active metal, never bare noble-metal fasteners in an active-metal sheet); (iv) apply coatings, and if a coating is used, coat the cathodic (more noble) member preferentially — a breakdown in a coating on the noble member exposes only a small cathodic area, whereas a breakdown in a coating on the anodic member exposes a small anode to the full cathodic demand of the bare noble metal and accelerates penetration; (v) add a sacrificial third metal (a galvanic anode, e.g. zinc or magnesium) deliberately more active than either member, so that it — not the structure — is consumed (this is cathodic protection, developed further in Question 6(b)); and (vi) where practical, keep the assembly dry or design for drainage so no continuous electrolyte film bridges the joint.

3.2 — (b) Sensitization and weld decay in stainless steels

How sensitization occurs. Austenitic stainless steels (e.g. Type 304/321, the alloy family of Question 2) rely on a minimum of roughly 10.5–12 wt% chromium in solid solution to sustain the passive $Cr_2O_3$-rich film. When such a steel is held, or slowly cooled through, the temperature range of roughly 425–870 °C (about 800–1600 °F) — exactly the range a weld heat-affected zone (HAZ) passes through as it cools from the fusion temperature — chromium diffuses to the austenite grain boundaries and precipitates there as chromium-rich $M_{23}C_6$ carbides. Because chromium diffuses far more slowly through the bulk austenite than carbon does, the carbide growth strips a zone immediately adjacent to each grain boundary of chromium, often depleting it below the 10.5–12% passivation threshold, while the grain interiors remain chromium-rich. The steel is now "sensitized": its bulk composition and even its bulk chromium content are unchanged, but a continuous, narrow, chromium-depleted zone now runs along every grain boundary in the HAZ. This is precisely why the attack is called weld decay: it appears as a band on each side of the weld, at the specific distance from the fusion line whose peak temperature during cooling matched the sensitizing range, not in the weld metal or the unaffected base metal.

Corrosion mechanism at the weld. The sensitized boundary zone and the chromium-rich grain interior now form a galvanic couple of exactly the kind described in part (a), on a microscopic scale: the depleted zone, unable to maintain a stable passive film, becomes anodic, while the still-passive grain interior is cathodic. Because the anodic path is a thin, continuous, low-chromium film along the boundary and the cathodic area (every grain face) is comparatively enormous, the area-ratio effect concentrates severe, highly localized attack along the grain boundaries — intergranular corrosion — which can proceed deep into the section with little visible surface loss and can drop entire grains out of the surface, catastrophically lowering strength and ductility in an otherwise sound-looking weld.

Engineering solutions. (i) Use a low-carbon grade (304L, 316L, carbon $\le0.03$ wt%): with less carbon available, far less chromium is needed to satisfy $M_{23}C_6$ formation, so the depleted zone never falls below the passivation threshold during a normal weld thermal cycle. (ii) Use a stabilized grade (321, containing titanium, or 347, containing niobium): Ti and Nb have a far stronger affinity for carbon than chromium does, so TiC/NbC precipitate preferentially throughout the grains (not preferentially at boundaries) during a post-weld stabilizing anneal, tying up the carbon before it can pull chromium out of solid solution at the boundaries — this is in fact why the paper's own worked figure in Question 2 specifies a Ti-stabilized "ANSI 321" alloy. (iii) Apply a solution anneal (heat to about 1040–1150 °C and quench) after welding, which redissolves the grain-boundary carbides and, by rapid cooling, does not give chromium time to diffuse back out and re-precipitate — effective but often impractical for large fabricated structures. (iv) Control the weld thermal cycle itself (lower heat input, interpass temperature control, or post-weld solution treatment of only the HAZ) to minimize the time spent in the 425–870 °C sensitizing range. (v) For service that will still see this temperature range for other reasons (e.g. high-temperature process piping), select a fully stabilized or low-carbon grade from the outset rather than relying on weld procedure alone.