Physical Constants and SI Units

Contents
  1. Mathematical constants
  2. The seven defining constants of the SI
  3. CODATA 2018 derived and measured constants
  4. Planck units
  5. SI base units

Since the revision that entered into force on 20 May 2019, the International System of Units (SI) is defined by assigning exact numerical values to seven defining constants of nature; every SI unit is then realised as a consequence of these fixed values [BIPM:2019]. All other physical constants are determined experimentally, and this treatise uses throughout the recommended values of the CODATA 2018 adjustment [Tiesinga:2021]. The present appendix collects, first, the seven defining constants with their exact values (The seven defining constants of the SI); second, the derived and measured constants that recur in the main text, quoted with their standard uncertainties (CODATA 2018 derived and measured constants); and third, the seven SI base units (SI base units).

Mathematical constants

Two dimensionless numbers precede every physical constant in this appendix, since they are not measured but proven: \(\ee = 2.718281828\ldots\) and \(\pi = 3.141592653\ldots\), both derived in full — \(\ee\) as the sum \(\sum 1/n!\) and the limit \(\lim(1+1/n)^n\) (Definition 7.53 and Proposition 7.57), \(\pi\) as twice the first positive zero of the cosine series and independently as the circle constant (Definition 7.61 and Proposition 7.62) — in Section 7.9.1 of Real Analysis. They are listed here only for numerical reference; their proofs are not repeated.

The digits are a separate claim from the existence of the constants, and worth separating explicitly. A definition that converges establishes that a number exists; it does not by itself justify writing down a single decimal. For \(\ee\) both steps are carried out: the truncation bound \(\frac{1}{(n+1)!}\le\ee-\sum_{k\le n}1/k!<\frac{1}{n\cdot n!}\) (Lemma 7.54) traps \(\ee\) between two explicit rationals, and twelve terms already force the nine decimals quoted above (Corollary 7.55). The trailing \(\ldots\) is a theorem too, not a typographical hedge: \(\ee\) is irrational (Proposition 7.56), so no finite decimal in this appendix equals it. Every entry below is a truncation; the only question is whether its error is stated.

The same two claims for \(\pi\) are asserted here but not yet derived.

Derivation pending.

Digits and irrationality of \(\pi\): an explicit error bound pinning \(\pi=3.141592653\ldots\), and Niven's proof that \(\pi\) is irrational. The latter needs \(\sin\pi=0\) and \(\cos\pi=-1\), hence the addition formulas, which the real-analysis chapter does not yet establish — it defines \(\pi\) as twice the first positive zero of the cosine series and proves it is the circle constant, but computes no decimal of it.

The seven defining constants of the SI

The 2019 revision fixes the numerical values listed in Table B.1 exactly, by definition and without uncertainty [BIPM:2019]. Each constant anchors one base unit, although the definitions are interlocking: the kilogram, for instance, is realised through the fixed value of the Planck constant together with the definitions of the second and the metre.

Defining constantSymbolExact valueAnchors
hyperfine transition frequency of $^{133}$Cs$\Delta\nu_{\mathrm{Cs}}$\(9192631770\,\mathrm{Hz}\)\(\mathrm{s}\)
speed of light in vacuum$c$\(299792458\,\mathrm{m}/\mathrm{s}\)\(\mathrm{m}\)
Planck constant$h$\(6.62607015\times 10^{-34}\,\mathrm{J}\,\mathrm{s}\)\(\mathrm{kg}\)
elementary charge$e$\(1.602176634\times 10^{-19}\,\mathrm{C}\)\(\mathrm{A}\)
Boltzmann constant$k$\(1.380649\times 10^{-23}\,\mathrm{J}/\mathrm{K}\)\(\mathrm{K}\)
Avogadro constant$N_{\mathrm{A}}$\(6.02214076\times 10^{23}\,/\mathrm{mol}\)\(\mathrm{mol}\)
luminous efficacy of \(540\times 10^{12}\,\mathrm{Hz}\) radiation$K_{\mathrm{cd}}$\(683\,\mathrm{lm}/\mathrm{W}\)\(\mathrm{cd}\)
The seven defining constants of the SI (2019 revision). All values are exact by definition [BIPM:2019].

CODATA 2018 derived and measured constants

Table B.2 lists the constants used most frequently in this treatise, with the CODATA 2018 recommended values and standard uncertainties [Tiesinga:2021]. Two entries deserve comment. The reduced Planck constant \(\hbar = h/2\pi\) is exact because \(h\) is exact; its decimal representation is merely truncated in the table. The standard acceleration of free fall \(g_0\) is not a constant of nature at all but a conventional exact value, adopted by the third Conférence Générale des Poids et Mesures in 1901 and retained in the SI Brochure [BIPM:2019]; the local acceleration of free fall differs from it at the level of a few parts in \(10^{3}\) and is what absolute gravimeters actually measure [Niebauer:1995].

QuantitySymbolValue$u_{\mathrm{r}}$
Newtonian constant of gravitation$G$\(6.67430(15)\times 10^{-11}\,\mathrm{m}^{3}/\mathrm{kg}/\mathrm{s}^{2}\)\(2.2\times 10^{-5}\)
reduced Planck constant$\hbar$\(1.054571817\times 10^{-34}\,\mathrm{J}\,\mathrm{s}\)exacta
electron mass$m_{\mathrm{e}}$\(9.1093837015(28)\times 10^{-31}\,\mathrm{kg}\)\(3.0\times 10^{-10}\)
proton mass$m_{\mathrm{p}}$\(1.67262192369(51)\times 10^{-27}\,\mathrm{kg}\)\(3.1\times 10^{-10}\)
vacuum magnetic permeability$\mu_0$\(1.25663706212(19)\times 10^{-6}\,\mathrm{N}/\mathrm{A}^{2}\)\(1.5\times 10^{-10}\)
vacuum electric permittivity$\epsilon_0$\(8.8541878128(13)\times 10^{-12}\,\mathrm{F}/\mathrm{m}\)\(1.5\times 10^{-10}\)
fine-structure constant$\alpha$\(7.2973525693(11)\times 10^{-3}\)\(1.5\times 10^{-10}\)
Bohr magneton$\mu_{\mathrm{B}}$\(9.2740100783(28)\times 10^{-24}\,\mathrm{J}/\mathrm{T}\)\(3.0\times 10^{-10}\)
standard acceleration of free fall$g_0$\(9.80665\,\mathrm{m}/\mathrm{s}^{2}\)exactb
Derived and measured constants, CODATA 2018 recommended values [Tiesinga:2021]. The column $u_{\mathrm{r}}$ gives the relative standard uncertainty.
a\,Exact because $\hbar = h/2\pi$ with $h$ fixed; the decimal digits shown are truncated. b\,Conventional exact value, not a constant of nature [BIPM:2019].

Planck units

The constants \(\hbar\), \(c\), and \(G\) together fix a length, mass, and time with no free numerical factor left undetermined by dimensional analysis alone.

Proposition B.1 (Planck units).

The unique (up to a dimensionless constant) combinations of \(\hbar\), \(c\), \(G\) with dimensions of length, mass, and time are

\begin{equation}\tag{B.1} \ell_{\mathrm P} = \sqrt{\frac{\hbar G}{c^{3}}}\ec\qquad m_{\mathrm P} = \sqrt{\frac{\hbar c}{G}}\ec\qquad t_{\mathrm P} = \sqrt{\frac{\hbar G}{c^{5}}} = \frac{\ell_{\mathrm P}}{c}\ep \end{equation}

Rests on Theorem 2.3 and Axiom 2.2.

Proof.

Derives Proposition B.1. Write \(\hbar^{a}c^{b}G^{d}\) and demand dimension \(\mathsf{L}\) (length). In SI, \([\hbar] = \mathsf{M}\mathsf{L}^{2}\mathsf{T}^{-1}\), \([c] = \mathsf{L}\mathsf{T}^{-1}\), \([G] = \mathsf{M}^{-1}\mathsf{L}^{3}\mathsf{T}^{-2}\). Matching powers of \(\mathsf{M}, \mathsf{L}, \mathsf{T}\) gives the linear system \(a - d = 0\) (\(\mathsf{M}\)), \(2a + b + 3d = 1\) (\(\mathsf{L}\)), \(-a - b - 2d = 0\) (\(\mathsf{T}\)); solving, \(a = d = 1/2\), \(b = -3/2\), i.e. \(\hbar^{1/2}c^{-3/2}G^{1/2} = \sqrt{\hbar G/c^{3}}\). By the Buckingham \(\pi\) theorem (Theorem 2.3), three quantities built from three dimensionally independent constants leave zero dimensionless combinations, so this exponent solution is unique and no undetermined dimensionless prefactor can be fixed by dimensional analysis alone — exactly the freedom absorbed into the convention that defines \(\ell_{\mathrm P}\) with prefactor \(1\). The mass and time forms follow by the same linear system with target dimension \(\mathsf{M}\) and \(\mathsf{T}\) respectively (or, more quickly, from \(m_{\mathrm P} = \hbar/(\ell_{\mathrm P} c)\) and \(t_{\mathrm P} = \ell_{\mathrm P}/c\), using \([\hbar] = \mathsf{M}\mathsf{L}^{2}\mathsf{T}^{-1}\) to fix the mass combination and dimensional consistency to fix the time combination).

Numerically, from the CODATA values of Table B.2,

\begin{equation}\tag{B.2} \ell_{\mathrm P} \approx 1.616\times 10^{-35}\,\mathrm{m}\ec\qquad m_{\mathrm P} \approx 2.176\times 10^{-8}\,\mathrm{kg}\ec\qquad t_{\mathrm P} \approx 5.391\times 10^{-44}\,\mathrm{s}\ec \end{equation}

the length and time scales at which quantum and gravitational effects are expected to become comparable in magnitude, and the mass scale against which the hierarchy problem of The Free Parameters of Physics is stated quantitatively.

SI base units

The seven base units of the SI, from which all coherent derived units are formed by products of powers, are collected in Table B.3 [BIPM:2019]. Every quantity in this treatise is expressed in these units or in coherent derived units built from them, in accordance with the axiom stated in the front matter.

Base quantityBase unitSymbol
timesecond\(\mathrm{s}\)
lengthmetre\(\mathrm{m}\)
masskilogram\(\mathrm{kg}\)
electric currentampere\(\mathrm{A}\)
thermodynamic temperaturekelvin\(\mathrm{K}\)
amount of substancemole\(\mathrm{mol}\)
luminous intensitycandela\(\mathrm{cd}\)
The seven SI base units [BIPM:2019].