Every equation in this book is written inside conventions that are set once and never mentioned again. Putting the units back takes exactly two constants.
🎯 Why this matters
A convention that is never restated is one you cannot check against. Every dimensional sanity check you would run on an engineering formula is unavailable until ħ and c are put back, which is how a wrong exponent survives into print unnoticed.Open this book at any page and you will find an equation like . Read it as an engineer and it is nonsense: you cannot add a momentum squared to a mass squared. Open it a page later and a particle has “a mass of 125 GeV” — an energy — and “a width of 2.5 GeV” — which is somehow a lifetime. Nothing here is sloppy. The whole book runs in a unit system where those statements are exact, and this page is where you learn it, because every other page assumes it.
This page is the toolkit: the unit convention, the two constants that undo it, the numbers from Appendices 1 and 2, and the symbol table for the whole site. Keep it open in a second tab.
The problem: three units for one thing
Write relativistic energy–momentum the way SI demands, and then again the way this book writes it:
The second line is not an approximation and it is not lazy. It is the statement that c and ħ are conversion factors, not physics — and that you may choose units in which they equal 1.
⚙️ Engineer’s bridge
You already do this. In DSP you stop writing frequencies in hertz and start writing them as , normalised to the sample rate — and suddenly the Nyquist limit is just instead of . In power engineering you switch to per-unit, and impedances stop carrying ohms. In both cases you picked a reference so that a recurring conversion factor becomes 1. Natural units natural units the convention ħ = c = 1, which collapses energy, mass and momentum into one unit (GeV) and makes time and length its inverse; the same move as normalising frequency by the sample rate in DSP. defined in the reference pages — open in glossary do exactly that with and .Where the analogy breaks: normalised frequency is dimensionless because you divided by something with the same dimension. Here you are declaring that seconds and metres were never independent dimensions to begin with — relativity says they are the same axis, so a conversion factor between them is as arbitrary as the factor between inches and feet.
Step 1: set c = 1
Special relativity says space and time are one geometry. If they are, measuring them in different units (metres and seconds) is a historical accident — like measuring a rectangle’s width in inches and its height in centimetres. Setting
declares one second and metres to be the same length. Immediately:
- velocities become dimensionless — a velocity is , a pure number between and , which is why the book writes everywhere;
- becomes , so a mass is an energy. The proton is “0.938 GeV”, full stop;
- momentum , energy and mass all carry the same unit, which is what makes legal.
Step 2: set ħ = 1
Quantum mechanics ties energy to frequency by and momentum to inverse wavelength by . Setting
declares an energy and an inverse time to be the same quantity. Now:
- a time is an inverse energy, so a lifetime and a width width Γ, the energy spread of an unstable state, tied to its lifetime by Γτ = ħ; literally the time–bandwidth relation, so a width quoted in MeV is an inverse lifetime. defined in the reference pages — open in glossary are the same information written two ways;
- a length is an inverse energy too (via ), so “1 GeV” is also a distance scale;
- angular momentum and action become pure numbers, which is why spins are quoted as , , rather than as .
💡 What this really says — one unit left in the whole subject, and everything is a power of it
After both steps there is exactly one unit left in all of particle physics, and by convention it is the electronvolt electronvolt the energy one elementary charge gains across 1 volt; the site's base energy unit, scaled as keV, MeV, GeV, TeV. defined in the reference pages — open in glossary (in practice GeV). Every quantity you will meet is some power of it. That is not a notational trick — it is the reason a single number, an energy, can answer “how heavy?”, “how short-lived?”, “how small?” and “how likely?” at once.Everything is a power of energy
| Quantity↕ | SI unit | Natural units↕ | Typical value in this book |
|---|---|---|---|
| Energy, mass, momentum | J, kg, kg m s⁻¹ | proton 0.938 GeV | |
| Time, length | s, m | proton radius 0.84 fm = 4.3 GeV⁻¹ | |
| Cross-section (area) | m² | pp total ≈ 110 mb | |
| Velocity | m s⁻¹ | dimensionless | β = v/c ∈ (−1, 1) |
| Angular momentum, action, spin | J s | dimensionless | spin 1/2, 1, 3/2 |
| s⁻¹ | GeV¹ | Z⁰ width 2.4952 GeV | |
| Force | N | GeV² | string tension ≈ 0.9 GeV² |
| Energy density | J m⁻³ | GeV⁴ | QCD vacuum ≈ (0.24 GeV)⁴ |
| Luminosity | cm⁻² s⁻¹ | GeV³ | LHC 2×10³⁴ cm⁻² s⁻¹ |
| Electric charge | C | e = √(4πα) ≈ 0.303 |
Undoing it: the only two constants you need
You will eventually want an answer in metres, seconds or barns. Two numbers do all of that work, and the book prints both in Appendix 2 as exact.
ħc turns an energy into a length
Exact by definition since the 2019 SI redefinition — ħ and c are both defined numbers.
Every symbol, one at a time
Hover or tap a symbol above — it lights up in the equation and its meaning, units and type appear here.
💡 What this really says — big energy means small distance, which is why every accelerator exists
Multiply any energy by nothing and read it as an inverse length. The number 197.3 MeV fm is the exchange rate. Because it is a product, big energy ⇒ small distance: the only way to see something small is to hit it hard. Every accelerator in this book exists because of that one line.🔢 Worked example — how hard must you hit a proton to see it?
The proton’s charge radius is fm. To resolve a structure of size you need a momentum transfer of at leastSo a few hundred MeV is the entry ticket to nuclear structure, and the GeV machines of Chapter 1 are what it takes to look inside the proton. Sanity check: 235 MeV is a quarter of the proton’s own mass — the right order, not a factor of a thousand out.
(ħc)² turns an inverse energy squared into a cross-section
Square Eq. (R.1) and convert fm² to barns — this is the constant that makes cross-sections computable.
Every symbol, one at a time
Hover or tap a symbol above — it lights up in the equation and its meaning, units and type appear here.
⚠️ Erratum in the printed book
Appendix 2 (p. 495) prints GeV² μbarn. Square the book’s own MeV fm and you get — the digits 372 and 327 are transposed in the table. The Particle Data Group value is GeV² mbarn, confirming it. This site uses the correct value everywhere, and it is a fair warning: a constants table is data, and data has typos. Cross-check constants against each other — must equal squared.🔢 Worked example — a real cross-section, end to end
The pointlike QED cross-section for (derived in §5.7) iswith and the squared centre-of-mass energy. At GeV, so GeV²:
That is the answer the theory gives, and it is in inverse GeV squared. Multiply by :
This is the famous "" that Chapter 6 divides by to define the ratio . You have just computed the denominator of one of the most important plots in the book, from first principles, in two lines.
Reproduce it
import numpy as np
hbar_c = 197.3269804e-3 # GeV fm (exact, App. 2)
hbar = 6.582119569e-25 # GeV s (exact, App. 2)
hbarc2 = hbar_c**2 / 100 * 1e6 # GeV^2 ub (1 b = 100 fm^2 = 1e6 ub)
alpha = 1 / 137.035999166
print(f"hbar*c = {hbar_c:.10g} GeV fm")
print(f"(hbar*c)^2 = {hbarc2:.10g} GeV^2 ub (book prints 389.3793271 - typo)")
print(f"Q to resolve 0.84 fm = {hbar_c / 0.84 * 1000:.1f} MeV")
sigma_gev2 = 4 * np.pi * alpha**2 / (3 * 100.0) # GeV^-2 at sqrt(s)=10 GeV
print(f"sigma(ee->mumu) at 10 GeV = {sigma_gev2 * hbarc2 * 1e3:.4f} nb") # 1 ub = 1e3 nb
print(f"Z0 lifetime from Gamma = {hbar / 2.4952:.4e} s")
print(f"proton mass = {0.938272088 * 1.7826619216e-27:.6e} kg")
print(f"300 K = {300 * 8.617333262e-14 * 1e12:.2f} meV") hbar*c = 0.1973269804 GeV fm (hbar*c)^2 = 389.3793719 GeV^2 ub (book prints 389.3793271 - typo) Q to resolve 0.84 fm = 234.9 MeV sigma(ee->mumu) at 10 GeV = 0.8685 nb Z0 lifetime from Gamma = 2.6379e-25 s proton mass = 1.672622e-27 kg 300 K = 25.85 meV
Try it: the converter
Everything above, as a knob. Type a number, pick a unit, and watch the same quantity in every other unit — including the natural-unit power of GeV. For an energy it also shows you the length, time and area that energy is.
⚖️ Natural-units converter
- eV1.00000 × 10⁹
- keV1.00000 × 10⁶
- MeV1000
- GeV1
- TeV0.001
- J1.60218 × 10⁻¹⁰
- kg1.78266 × 10⁻²⁷
- K1.16045 × 10¹³
…and the same number read as a length, a time and an area
- ħc / E = 0.197327 fm— the smallest distance this energy can resolve
- ħ / E = 6.58212 × 10⁻²⁵ s— the lifetime of a state of that width
- (ħc)² / E² = 389.379 μb— the cross-section scale it sets
ħ = c = 1 makes E, m and p one dimension. Back to SI: m = E/c², T = E/k.
Appendix 2 — the constants themselves
Transcribed from p. 495. Click any cell with a dotted underline for what the constant is and where the book uses it. Note the exact column: since the 2019 SI redefinition, several of these are not measurements at all.
| Quantity↕ | Symbol | Value | Uncertainty (ppb)↕ |
|---|---|---|---|
| Speed of light in vacuum | exact | ||
| Planck constant | exact | ||
| Planck constant, reduced | exact | ||
| Conversion constant | exact | ||
| Conversion constant | exact | ||
| Elementary charge | exact | ||
| Electron mass | 9.109 383 7015(28) × 10⁻³¹ kg = 0.510 998 950 00 MeV | 0.30 | |
| Proton mass | 1.672 621 923 69(51) × 10⁻²⁷ kg = 938.272 088 16 MeV | 0.31 | |
| 5.788 381 8060(17) × 10⁻¹¹ MeV T⁻¹ | 0.30 | ||
| 3.152 451 258 44(96) × 10⁻¹⁴ MeV T⁻¹ | 0.31 | ||
| 0.529 177 210 903(80) × 10⁻¹⁰ m | 0.15 | ||
| 1 / fine structure constant | 0.11 | ||
| Newton constant | 6.674 30(15) × 10⁻¹¹ m³ kg⁻¹ s⁻² | ||
| Fermi constant | 510 | ||
| Weak mixing angle | 1.7 × 10⁵ | ||
| Strong coupling constant | 7.6 × 10⁶ | ||
| Avogadro number | 6.022 140 76 × 10²³ mol⁻¹ | exact | |
| Boltzmann constant | exact |
Values from CODATA 2018 (Tiesinga <em>et al.</em>, Rev. Mod. Phys. <strong>93</strong> 025010, 2021); α from Fan <em>et al.</em> (2023); G<sub>F</sub>, α<sub>s</sub> and sin²θ<sub>W</sub> from the Particle Data Group (Workman <em>et al.</em> 2022). Uncertainties are one standard deviation on the last digits, in parts per 10⁹.
⚙️ Engineer’s bridge — why “exact” is not a brag
Seven entries above are exact. Not “measured very precisely” — exact, with no uncertainty at all, because in 2019 the SI was redefined to fix c, h, q_e, N_A and k at chosen numerical values and derive the metre, kilogram, ampere, mole and kelvin from them.This is the same move as choosing a fixed-point format. You do not measure how many volts one LSB is; you define the full-scale reference and everything else is counted in units of it. The physics did not change in 2019; the direction of the definition did. And it is why can be quoted to unlimited digits while the proton mass cannot: one is arithmetic, the other is an experiment.
Where it breaks: “exact by definition” and “known to unlimited digits” are not the same claim, and the SI redefinition blurs them. is exact because and are fixed numbers — but a mass expressed in kilograms is now traceable to those same fixed constants through an apparatus with a real uncertainty, so exactness in the definition does not propagate into the measurement. What became arithmetic is the conversion; the weighing is still an experiment, and the proton’s mass is uncertain at for reasons no definition can remove.
Orders of magnitude — the map you should carry
Every energy in this book is somewhere on this line, and by ħc every energy is also a distance. Read it in both directions.
The symbols this book uses
Particle physics ran out of letters decades ago. This table is the site-wide convention; every chapter obeys it.
| Symbol | Means↕ | Group↕ | Watch out for |
|---|---|---|---|
| Electric charge, in units of the elementary charge | flavour | ±1/3, ±2/3 for quarks; always an integer for hadrons | |
| Strong isospin and its third component | flavour | An SU(2) label over (u, d) only — not the weak isospin below | |
| Strangeness, charm, beauty, top | flavour | Sign convention: the s quark has S = −1, the b quark B = −1 | |
| Baryon number, 1/3 per quark | flavour | Written B in the book — do not confuse with beauty B | |
| Strong hypercharge, Y = 𝓑 + S + C + B + T | flavour | One axis of every SU(3) weight diagram | |
| Spin and parity of a state | state | e.g. the pion is 0⁻, the ρ is 1⁻ | |
| Isospin and G-parity (mesons only) | state | G-parity is C combined with an isospin rotation (§3.10) | |
| Weak isospin, its third component, weak hypercharge | electroweak | Left- and right-handed components of the same fermion differ — that is the whole point | |
| Z⁰ coupling factor, with s² ≡ sin²θ_W | electroweak | Ch. 9 uses it constantly | |
| Electromagnetic and strong coupling constants | dynamics | Both run with Q²; always ask “at what scale?” | |
| Fermi constant | dynamics | Units GeV⁻² — a contact interaction | |
| Width and lifetime, Γτ = ħ | dynamics | Quoted interchangeably; a width in MeV is an inverse lifetime | |
| Cross-section | dynamics | Also a Pauli matrix, also a Σ hyperon, also a standard deviation | |
| v/c and the Lorentz factor (1−β²)^(−1/2) | kinematics | γ is also the photon and also a Dirac matrix γ^μ | |
| Centre-of-mass energy | kinematics | s is a Mandelstam invariant (§1.4), not a strange quark here | |
| Neutrino mass eigenstates | neutrino | Distinct from the flavour eigenstates ν_e, ν_μ, ν_τ — Ch. 10 lives on that difference | |
| Antiparticles carry an overbar | convention | On this site they are drawn outlined, never filled |
Greek letters, and the overloading problem
Appendix 1 is just the alphabet. The useful version is knowing which glyph is overloaded — because the book will not warn you.
| Lower | Upper | Name↕ | Used in this book for |
|---|---|---|---|
| α | Α | alpha | fine-structure constant; α_s; an α particle (⁴He nucleus) |
| β | Β | beta | v/c; β decay |
| γ | Γ | gamma | photon; Lorentz factor; Dirac matrices γ^μ — and uppercase Γ is a decay width |
| δ | Δ | delta | CP-violating phase δ; uppercase Δ is the 3/2 baryon and also “difference” (Δm²) |
| ε | Ε | epsilon | CP-violation parameters ε, ε′; polarization vector |
| ζ | Ζ | zeta | rare — but Z⁰ (Latin Z) is everywhere |
| η | Η | eta | the η and η′ mesons |
| θ, ϑ | Θ | theta | scattering angle; θ_W weak mixing angle; θ_C Cabibbo; θ₁₂, θ₁₃, θ₂₃ neutrino mixing |
| ι | Ι | iota | rare |
| κ | Κ | kappa | anomalous coupling parameters (Ch. 9) |
| λ | Λ | lambda | wavelength; Gell-Mann matrices λ_a; Wolfenstein λ — uppercase Λ is the hyperon and Λ_QCD |
| μ | Μ | mu | the muon; a magnetic moment; the prefix micro (μb) |
| ν | Ν | nu | neutrino (always, in this book) |
| ξ | Ξ | xi | uppercase Ξ is the cascade hyperon |
| ο | Ο | omicron | not used — too close to zero |
| π | Π | pi | the pion; and 3.14159… |
| ρ | Ρ | rho | the ρ meson; a density; Wolfenstein ρ̄ |
| σ, ς | Σ | sigma | cross-section; Pauli matrices; a standard deviation — uppercase Σ is the hyperon and a sum |
| τ | Τ | tau | the τ lepton; a lifetime |
| υ | Υ, ϒ | upsilon | uppercase ϒ is the bb̄ bottomonium state |
| ϕ, φ | Φ | phi | the ϕ meson; an azimuthal angle; a scalar field |
| χ | Χ | chi | the χ_c charmonium states; chirality; a χ² fit statistic |
| ψ | Ψ | psi | the J/ψ and ψ′; a wave function |
| ω | Ω | omega | the ω meson; an angular frequency — uppercase Ω is the Ω⁻ baryon and a solid angle |
🔑 If you remember only three things
-
A width is a lifetime. 2.5 GeV is not an energy spread to widen a cut around — it is a particle that lives s.
-
GeV is an area, not a slip. A cross-section carries two inverse powers of energy, and the factor that turns it into barns is where is hiding.
-
Powers of energy do not tell you which quantity you have. A mass, a momentum, a temperature and an inverse length all read as GeV, so the units can never catch a mistake the way they would elsewhere.
Where this is used
- §1.6 Measurement Units applies all of this to real collisions and decays, and is where barns barn the natural area unit for cross-sections, 1 b = 10⁻²⁸ m² = 100 fm²; sub-multiples mb, μb, nb, pb, fb appear throughout. defined in the reference pages — open in glossary first earn their keep.
- §1.8 Scattering Experiments turns a cross-section into an event rate — the “effective target area” picture.
- The particle data explorer is the same Appendix 3 tables, made searchable; every mass there is in the units defined here.
✅ Check yourself — units and notation
0/5 answered · 0 correct
1.A paper reports a new resonance with a width MeV. Roughly how long does it live?
Hint: MeV s.
2.Why can be quoted as exactly MeV fm, while the proton mass carries an uncertainty?
3.In the converter above, set the dimension to Energy and type 1 GeV. Which of the following will it show?
4.A theorist hands you a cross-section as the pure number GeV. What is it in barns?
Hint: GeV² μbarn.
5.The book's Appendix 2 prints GeV² μbarn, while this page uses . What should you conclude?