+Avogadro's number changed (and AMU with that)

[timbrisc]

Issue migrated from trac ticket # 5775

component: unit definitions | priority: major

2020-09-16 17:48:55: [email protected] created the issue

---

Originally from Lynch, Paul (NIH/NLM/LHC) [E] [email protected] for Clem McDonald:

A needed change to a couple of UCUM constants:

In 2019 the definition of “mole” was revised (https://en.wikipedia.org/wiki/2019_redefinition_of_the_SI_base_units#Mole) to be exactly 6.02214076×10^23^ (instead of being experimentally determined), but UCUM has an older value of 6.0221367×10^23^.

Also, the conversion of an atomic mass unit to g (experimentally determined) needs an update. Per the 2018 CODATA recommendations (https://physics.nist.gov/cuu/pdf/wallet_2018.pdf) the recommended value is now 1.660 539 066 60(50) × 10^−24^ g. UCUM has 1.6605402 × 10^-24^. Note that the “(50)” is the uncertainty on the previous two digits and should not be included in the value.

The re-definition of a mole as a number affects how conversions between mass and moles are done, because now molecular weight is no longer exactly equal numerically (modulo units) to molar weight, though the difference is very slight (about the 7th significant digit). However, I don’t think that affects any part of the text in the specification, as far as I can tell.

So, there is the definition of the two constants to change (“Table 4: SI units”, and “Table 5: Other units…”), and also there are several units in “Table 26: Example Unit Terms by Term” which involve “mol” whose “canonical form value” will need to be revised.

I don’t know how updates to constants have been handled in the past, but perhaps it is worth considering incrementing the version number of the standard from 2.1 to 2.1.1.