Docs / What we have found

A nuclide is named, or numbered, as something else

Part of What we have found, the class where the nuclide a flow names is not the nuclide its factors were computed for — or is not a nuclide at all.

Palladium-234m does not exist

EF 3.1 ships twelve flows named Palladium-234m, all in becquerels, in radiological release compartments — ground-level air, aircraft cruise height, high stack, ocean, surface water, agricultural ground.

Palladium is element 46 and its heaviest known isotope is around mass 128. There is no Pd-234, so there is no Pd-234m. The flow is protactinium-234m — Pa-234m, the short-lived isomer in the uranium-238 decay chain, and one of the nuclides an inventory of nuclear power is most likely to report. The element symbol Pa was expanded to the wrong element's name.

EF half-knows. The general comment on all twelve rows reads:

CAS No and Synonyms removed; were erroneous. Correct CAS to be identified. Possibly has none assigned.

The identifiers were withdrawn because they did not match the name. The name that made them not match was left in place, and the flow arrives carrying no registry number at all — so nothing but the name says what it is, and the name is the thing that is wrong.

Where the identifiers are weakest

The name was left to do the identifying because nothing else could.

A radionuclide flow is measured in becquerels, and what it is depends on two things a chemical flow does not have: which nuclide, and which nuclear state. Neither is expressible as a registry number in practice, so nuclide rows are where the lists' identifiers are weakest — and where a name is doing work no number is checking.

The same shape, four more times

The silver case is worth following, because the name is wrong in every list at once — a shared vendor convention, not one list's slip, so it cannot be settled by preferring the other list. Half-life alone does not settle it either: 49 of the 81 matched nuclides in this list have a half-life under a year and most are legitimate. What settles it is that EF states which nuclide it means a second time, in its own characterisation factor. The four water rows carry Ionising radiation, human health at 0.0236 kBq U235-eq per kBq. Of the twenty-one nuclides carrying a water factor under that method, not one is short-lived: the band runs from Iodine-129 down to Strontium-90, and this value sits between Antimony-124 (60.2 days) and Manganese-54 (312 days). An exposure efficiency of that size needs a nuclide that survives long enough to be ingested. A 24.6-second beta emitter discharged to the ocean does not. The factor was computed for the isomer (#219).

The decisions. Palladium-234m is renamed, and the corrected label is what lets the nuclide be recognised at all — typed as an isotope and a radionuclide, where before it was untyped for having neither structure nor registry identifier. Silver-110 is renamed to Silver-110m and keeps Silver-110 as an alternative label, which is the opposite of the trichloroethane and paraquat decisions and deliberately so: it is not another substance's name, it is the name both vendor lists ship, and deleting it would remove the only string a consumer holding one of those inventories could search for. Uranium-238 gets its own registry number on every list that carries it.

What is still wrong. Thorium-232 and Praseodym-147 keep the element's registry number, because no cited nuclide-specific number exists for them and inventing one would be worse than publishing the vendor's. A consumer joining on CAS will re-merge those nuclides with their elements — which is exactly what the identifiers say, and exactly what the nuclide identity in this list says they are not (#200, #241).