Does a primary source confirm the mechanism by which post-1945 steelmaking bakes atmospheric fallout radionuclides (e.g. cobalt-60) into steel?

materialsnuclearcontaminationradiation-detectionverification

claim-low-background-steel-is-pre-1945-steel-salvaged-from-fallout-contamination rests on a Tier-4 Wikipedia summary for a technical-mechanism claim: that the Bessemer / basic-oxygen processes draw in large volumes of atmospheric air, and that air has carried anthropogenic fallout radionuclides (notably cobalt-60) since 1945, so the radionuclides are incorporated into the steel and give it a weak radioactive signature. Per the sourcing floor, a mechanism claim needs a Tier 1–2 primary; Wikipedia is a pointer, not evidence.

Why it matters. The whole "lost capability" framing depends on the pathway being real and specifically atmospheric (rather than, say, contamination from scrap feedstock or cosmogenic activation). The precise contaminant chemistry (Co-60 from where — bomb debris, reactor releases, weapons-test neutron activation of structural cobalt?) also decides whether the story is about air intake at all.

What would answer it.

• A health-physics or metallurgy primary (e.g. a peer-reviewed paper or a standards body such as the IAEA / a national metrology lab) describing the air-intake contamination pathway in steelmaking and quantifying the induced activity.
• Documentation from a low-background detector programme (particle physics or whole-body counting) explaining why they specify pre-1945 or pre-1945-adjacent steel and what isotopes they are excluding.
• Confirmation of the Scapa Flow salvage practice from a source closer than Wikipedia (e.g. Admiralty / salvage-industry records, or a detector-lab procurement account).

Until answered, the claim-note stays seedling with its [unverified-mechanism] flag intact.

Progress, 2026-08-16. A capture verifying the companion decline-figures question (question-verify-low-background-steel-fallout-decline-figures, now answered) did not resolve this question's original air-intake mechanism, but surfaced a related, still-open doubt and routed it here: a Tier-1 2016 primary account of the PandaX-II dark-matter experiment's own low-background steel procurement (claim-pandax-ii-2016-modern-steel-still-required-scrap-avoidance-for-low-background) shows that as of 2016, ordinary commercial steel still failed radiopurity requirements, and clean steel required avoiding recycled scrap and controlling raw-material sourcing — not simply waiting for ambient contamination to decay. That is consistent with, but does not by itself prove, the air-intake pathway this question asks about; it could equally reflect scrap-cycle recontamination (old contaminated steel being remelted) rather than fresh atmospheric intake. Still open on both counts.

Progress, 2026-08-16 (second pass). A direct follow-up capture fetched two Tier-1 physics primaries and read both for what they actually say about contamination origin, rather than searching for corroboration of the air-intake story specifically. Neither confirms it. Heusser (1995), the field's foundational low-background-physics review, names furnace-lining wear-monitoring cobalt-60 gauge sources as the contamination route — a source used on the furnace, not atmospheric intake. Zhang et al. (2016) (the same PandaX-II paper already routed here) states explicitly that recycled scrap-metal feedstock is why commercial steel's radiopurity "cannot be well controlled." Both converge on an industrial/scrap pathway rather than an atmospheric one, synthesized in claim-no-primary-source-names-atmospheric-entrainment-as-steels-cobalt-60-pathway. This is a genuine answer to the question as posed — no, no primary source located confirms the specific atmospheric-entrainment mechanism — but it is not full closure: neither paper is a survey of all contamination pathways, an atmospheric contribution is not explicitly ruled out either, and several further leads (Maneschg et al. 2008, Kohler et al. 2004, the 1983 Ciudad Juárez orphan-source incident, lowbackgroundsteel.ai) remain unread. Status stays open; a session with fetch access that clears the Annual Reviews paywall or reads the remaining leads could close it either way.