# Mohr:2025 — CODATA 2022 recommended values

**Citation.** P. J. Mohr, D. B. Newell, B. N. Taylor and E. Tiesinga,
"CODATA Recommended Values of the Fundamental Physical Constants: 2022",
*Reviews of Modern Physics* (CODATA 2022 adjustment).

**Access.** Open. NIST publishes the complete listing at
<https://physics.nist.gov/cuu/Constants/> and the plain-text table at
<https://physics.nist.gov/cuu/Constants/Table/allascii.txt>.

## What this reference supports in the treatise

The modern value that an experiment chapter compares its historical
measurement against. Cited by the Cavendish, Joule, Franck–Hertz, positron,
electron *g*−2, precision-spectroscopy, quantum-Hall and superconductivity
experiment chapters, among others.

`../data/Mohr_2025.csv` holds the 34 constants those chapters quote, so a
claim can be diffed against the source without reopening the paper.

## Provenance

- **Fetched, not transcribed by hand.** `tools/bibcheck/build_codata.py`
  downloads NIST's own ASCII listing and reads the fixed-width columns it
  documents. The only edit is removing the spaces NIST uses as digit
  separators (`1.602 176 634 e-19` → `1.602176634e-19`).
- Values are **recommended values from a least-squares adjustment** of the
  world's measurements, not the result of a single experiment. A chapter
  must not present one as "the measured value" of its own experiment.
- `(exact)` marks a constant fixed by the 2019 SI redefinition — *c*, *h*,
  *e*, *k*, *N*_A, and the derived von Klitzing and Josephson constants.
  These have no uncertainty *by definition*, which is itself the point the
  quantum-Hall and superconductivity chapters make.
- NIST prints some exact values with a trailing `...` (the decimal does not
  terminate); the CSV keeps that notation rather than silently truncating.

## Caveats

- Supersedes CODATA 2018 (`Tiesinga:2021`). Where a chapter quotes the 2018
  adjustment, check it against `../data/Tiesinga_2021.csv` instead — the
  two differ in the last digits and mixing them would misstate a
  precision claim.
- The fine-structure constant is the case to watch: the Cs (`Parker:2018`)
  and Rb (`Morel:2020`) determinations disagree at more than 5σ, and the
  CODATA value is an adjustment over both. The *g*−2 chapter must say which
  input it means rather than quoting the adjusted value as if the
  disagreement were settled.