Corrosion of metals

"Iron rusts in air" is not credited: it never names oxygen specifically, and it omits water entirely. Naming only water is equally incomplete. The mark-scheme answer is that oxygen and water must both be present. Each is necessary and neither alone is sufficient — remove either and no rust forms. Nitrogen takes no part, so it is never a condition to state.

Asked why galvanised steel resists rust where the coating is scratched, candidates restate the barrier idea — "the zinc keeps oxygen and water out". That holds only while the coating is intact. The scratched case needs sacrificial protection: zinc is more reactive, so at the exposed metal zinc is oxidised instead of iron. Paint has no such backup.

"More reactive" here does not mean fizzing harder with acid or burning more fiercely. For sacrificial protection it means the metal loses electrons more readily to form its positive ion: zinc forms $\text{Zn}^{2+}$ more readily than iron forms $\text{Fe}^{2+}$ or $\text{Fe}^{3+}$, so zinc is oxidised in preference. Oxidation is loss of electrons; reduction is gain.

"The paint stops the rust" describes the outcome, not the mechanism, and scores nothing on a describe how question. The required chain is: the coating forms a continuous physical layer over the iron, this excludes oxygen and water, and without both rusting cannot occur. Name the layer, name what it keeps out, then link back to the conditions.

In the three-test-tube experiment every tube must remove exactly one condition. Candidates draw the deoxygenated tube without the layer of oil sealing it, so air redissolves, and label the drying agent vaguely instead of naming anhydrous calcium chloride (or silica gel) removing water vapour. Without those labels no tube isolates a single variable.