Antiparticles & the CPT picture

Antimatter is the one prediction of twentieth-century physics that arrived before any experiment asked for it. In 1928 Dirac wrote down the relativistic wave equation for the electron, and the equation would not behave: alongside the ordinary positive-energy electron it insisted on a second family of solutions with energy running down to minus infinity. There is no way to throw these solutions away — they are needed to make the set complete — so Dirac was forced to ask what they mean. His answer, sharpened over the next three decades into the modern picture, is that every charged fermion is shadowed by an antiparticle of exactly the same mass and opposite charge, and that the two are locked together by the deepest symmetry we know of, CPT. This article is the synthesis of the antimatter head. It takes the relativistic energy relation $E=\pm\sqrt{p^2c^2+m^2c^4}$ from the kinematics article and the Dirac equation from its own, and asks what the negative root actually is. Three answers stack up, each superseding the last. Dirac's hole theory fills the negative sea and calls a hole a positron. The Feynman–Stückelberg interpretation drops the sea and reads a negative-energy solution running backward in time as a positive-energy antiparticle running forward. And quantum field theory makes it exact: a single field operator creates particles and annihilates antiparticles, while the CPT theorem — Lorentz invariance plus locality plus a Hermitian Hamiltonian — guarantees particle and antiparticle share mass and lifetime to all orders. The positron $e^+$ is the archetype; the same logic runs through every fermion in the Standard Model.

Key takeaways

Relativistic quantum mechanics forces a doubling: the relation $E=\pm\sqrt{p^2c^2+m^2c^4}$ has a negative root that cannot be discarded, and it becomes the antiparticle.
For every charged fermion there is an antiparticle of identical mass and opposite charge (and opposite additive quantum numbers): the anti-electron is the positron $e^+$ , the anti-up-quark is $\bar u$ .
Three views layer up: Dirac's filled negative-energy sea (a hole = a positron), the Feynman–Stückelberg reinterpretation (negative energy backward in time = positive energy forward), and the field-theory operator that ties both together.
The link is enforced by the CPT theorem: any local, Lorentz-invariant, Hermitian quantum field theory is CPT-invariant, so $m_a=m_{\bar a}$ and $\tau_a=\tau_{\bar a}$ exactly.
CPT is the most precisely tested prediction in physics — the neutral-kaon mass split is below $10^{-18}$ , and $e^+$ / $e^-$ $g$ -factors agree to $\sim10^{-12}$ .
Key idea: marrying quantum mechanics to special relativity forces a second family of states — every fermion is shadowed by an antiparticle of equal mass and opposite charge, and CPT makes that pairing exact.

The core insight

Requiring a quantum theory consistent with special relativity forces a doubling of states: every fermion is shadowed by an antiparticle of exactly equal mass and opposite charge. Antimatter is not an optional add-on — it is demanded by the negative root of $E^2=p^2c^2+m^2c^4$ , and the CPT theorem welds the pair together.

Where this lives today

Antimatter has moved from prediction to precision laboratory. At CERN's Antiproton Decelerator, the BASE experiment traps single antiprotons and, in 2022, compared the antiproton-to-proton charge-to-mass ratio to 16 parts per trillion — the sharpest baryonic CPT test yet, consistent with exact equality. Next door, the ALPHA collaboration confines antihydrogen (an antiproton bound to a positron): its 2018 measurement of the $1S\!\to\!2S$ transition matched ordinary hydrogen to $2\times10^{-12}$ , and in 2023 ALPHA-g watched antihydrogen fall down under gravity, ruling out antigravity. Meanwhile the most consequential antimatter question is cosmological and wide open: if CPT makes matter and antimatter so symmetric, why is the observed universe made almost entirely of matter? Answering it requires CP violation beyond what the Standard Model supplies — the subject of baryogenesis. On the applied side, positron annihilation is a daily clinical tool: PET imaging maps metabolism from the back-to-back 511 keV photons of $e^+e^-$ annihilation in the body.

The physical picture

Start with the fact that will not go away. The relativistic energy expression is quadratic, $E^2=p^2c^2+m^2c^4$ , so for every momentum there are two roots, $E=+\sqrt{\dots}$ and $E=-\sqrt{\dots}$ , separated by a forbidden gap of width $2mc^2$ . In classical mechanics you keep the positive root and forget the other; in quantum mechanics you cannot, because the transition amplitudes connect the two families and a wavefunction built only from positive-energy pieces is not complete. Dirac's first response was to imagine the negative-energy states already filled — an infinite sea of electrons that the Pauli principle keeps ordinary matter from falling into. Lift one sea electron into the positive continuum with a photon of at least $2mc^2$ and you leave behind a hole: an absence of negative charge and negative energy that looks for all the world like a particle of positive charge and positive energy. That hole is the positron, and the first figure is that spectrum.

Figure

The Dirac energy spectrum: two continua of solutions E = ±√(p²c² + m²c⁴) separated by a forbidden gap 2mc². For the electron the gap is 2mₑc² = 1.022 MeV, so a photon of at least that energy lifts a sea electron to the positive continuum and leaves a hole — a positron. This is pair creation γ → e⁺ + e⁻

The sea is a picturesque crutch, and it was known to be dispensable by 1934, when Pauli and Weisskopf built a consistent theory of charged spin-zero bosons — which cannot form a Pauli sea at all — that still came with antiparticles. The durable statement is the Feynman–Stückelberg one. A negative-energy solution carries the time factor $e^{-iEt}=e^{+i|E|t}$ ; read literally this is a positive-energy object whose clock runs the other way. Rather than an electron of energy $-|E|$ crawling forward in time, we read it as a positron of energy $+|E|$ moving forward — equivalently an electron moving backward. In a Feynman diagram the two are one line: the amplitude to emit an antiparticle of four-momentum $p$ equals the amplitude to absorb a particle of four-momentum $-p$ . The second figure is the cleanest case — an electron and positron annihilating into two photons — drawn so the positron's charge-arrow points against the flow of time.

Figure

Feynman–Stückelberg picture of e⁺e⁻ annihilation at rest: time runs upward. The electron line carries its charge-arrow forward in time; the positron is the same line with its arrow reversed — a charge-conjugated electron running backward. The pair annihilates into two back-to-back photons, each carrying Eγ = mₑc² = 0.511 MeV

The core equations

Everything begins with the relativistic dispersion relation, which the special-relativity article owns. It is quadratic in $E$ , and the quantum theory is obliged to keep both roots, so the negative branch is present from the first line.

E^{2}=|\mathbf{p}|^{2}c^{2}+m^{2}c^{4}\quad\Longrightarrow\quad E=\pm\sqrt{|\mathbf{p}|^{2}c^{2}+m^{2}c^{4}} \qquad (1)

where: $E$ — total relativistic energy; $\mathbf{p}$ — three-momentum; $m$ — rest mass; $c$ — speed of light. The positive root is the ordinary particle; the negative root, once completeness forbids discarding it, is what the rest of this article decodes. The Dirac equation is the relativistic wave equation that keeps the probability density positive — its virtue over Klein–Gordon — yet it inherits the negative root all the same. Being first order in derivatives is exactly why it needs four-component spinors and the gamma matrices.

(i\gamma^{\mu}\partial_{\mu}-m)\psi=0,\qquad \{\gamma^{\mu},\gamma^{\nu}\}=2g^{\mu\nu}\mathbb{1} \qquad (2)

where: $\gamma^{\mu}$ — Dirac matrices obeying the Clifford algebra; $\psi$ — four-component spinor field; $m$ — mass (natural units $\hbar=c=1$ ); $g^{\mu\nu}$ — Minkowski metric. Squaring the operator recovers $E^2=p^2+m^2$ , so both energy signs survive: linearization changes the number of components, not the spectrum. Charge conjugation is the operation that turns a solution describing an electron in a field $A_\mu$ into one describing a positron in the same field. Applying it exposes the opposite-charge partner explicitly and shows the mass is untouched.

\psi^{c}=C\bar\psi^{\,T}=i\gamma^{2}\psi^{*},\qquad \big[i\gamma^{\mu}(\partial_{\mu}+ieA_{\mu})-m\big]\psi=0\ \longrightarrow\ \big[i\gamma^{\mu}(\partial_{\mu}-ieA_{\mu})-m\big]\psi^{c}=0 \qquad (3)

where: $\psi^{c}$ — charge-conjugate spinor; $C=i\gamma^{2}\gamma^{0}$ — charge-conjugation matrix (in the Dirac basis $\psi^c=i\gamma^2\psi^*$ , up to a convention phase); $e$ — proton charge. The sign of the electromagnetic coupling flips ( $+ieA_\mu\to-ieA_\mu$ ) while $m$ stays put: same mass, opposite charge — the definition of an antiparticle. Quantum field theory removes the last trace of the sea. A single Dirac field is a sum of two operators: one that annihilates a particle and one that creates an antiparticle, each with its own spinor and its own energy sign built into the exponential.

\psi(x)=\int\!\frac{d^{3}p}{(2\pi)^{3}}\,\frac{1}{\sqrt{2E_{\mathbf{p}}}}\sum_{s}\Big(a_{\mathbf{p}}^{s}\,u^{s}(p)\,e^{-ip\cdot x}+b_{\mathbf{p}}^{s\dagger}\,v^{s}(p)\,e^{+ip\cdot x}\Big) \qquad (4)

where: $a_{\mathbf{p}}^{s}$ — annihilates a particle of momentum $\mathbf{p}$ , spin $s$ ; $b_{\mathbf{p}}^{s\dagger}$ — creates an antiparticle; $u^s,v^s$ — positive- and negative-frequency spinors; $E_{\mathbf p}=+\sqrt{p^2+m^2}>0$ . Both terms carry positive energy: the old negative root now sits in the $e^{+ip\cdot x}$ factor of the creation operator, so no sea is required and the vacuum is simply the state killed by every $a$ and $b$ . Finally the theorem that binds particle to antiparticle. Let $\Theta=CPT$ be the combined charge-conjugation, parity, and time-reversal operator. Any local, Lorentz-invariant theory with a Hermitian Hamiltonian commutes with $\Theta$ , and the consequences are exact and non-negotiable.

\Theta=CPT,\qquad \Theta\,\mathcal{H}\,\Theta^{-1}=\mathcal{H}\ \Longrightarrow\ m_{a}=m_{\bar a},\quad \tau_{a}=\tau_{\bar a},\quad Q_{a}=-Q_{\bar a},\quad \boldsymbol{\mu}_{a}=-\boldsymbol{\mu}_{\bar a} \qquad (5)

where: $m,\tau,Q,\boldsymbol\mu$ — mass, total lifetime, charge, magnetic moment; subscripts $a,\bar a$ — particle and antiparticle. CPT does not require $C$ , $P$ , or $T$ to hold separately — each is violated in nature — only their product, which is why a measured mass or lifetime difference between a particle and its antiparticle would break physics at its foundations.

Derivation AWhy the negative-energy states survive the Dirac linearization

The goal is to trace exactly where the antiparticle enters, showing that Dirac's equation cures the disease that killed the Klein–Gordon single-particle picture — negative probability density — without curing the negative energies themselves, so the second family of solutions is a permanent feature demanding physical meaning, not a defect to engineer away. We compare the two equations at the continuity equation and the plane-wave spectrum, and count solutions.

Step 1. The Klein–Gordon equation $(\partial^2+m^2)\phi=0$ is second order in time, so specifying $\phi$ at one instant does not fix its evolution — you also need $\dot\phi$ — and the conserved density that follows from it, $\rho\propto i(\phi^*\dot\phi-\dot\phi^*\phi)$ , is not positive definite. Why: a density that can go negative cannot be a single-particle probability, which is the flaw that first made negative-energy solutions look fatal.

Step 2. Dirac demands an equation first order in $\partial_t$ , $i\partial_t\psi=(\boldsymbol\alpha\cdot\mathbf p+\beta m)\psi$ , forcing $\boldsymbol\alpha,\beta$ to be matrices with $\alpha_i\alpha_j+\alpha_j\alpha_i=2\delta_{ij}$ and $\beta^2=1$ . The conserved density is now $\rho=\psi^\dagger\psi\ge0$ , positive definite. Why: first order in time gives a bona fide positive probability, which is the property Klein–Gordon lacked and the reason Dirac's equation is the correct one-particle starting point.

Step 3. But squaring the Dirac operator returns each component to $(\partial^2+m^2)\psi=0$ , so the plane-wave spectrum is unchanged: $E=\pm\sqrt{p^2+m^2}$ , both signs present. For each momentum there are four independent spinors — two spin states at $E>0$ and two at $E<0$ . Why: the negative energies are not an artifact of a bad choice; they are locked in by relativity itself, so a complete set of states must include them.

Step 4. The completeness relation over all four spinors shows that dropping the $v$ -spinors leaves the basis incomplete, so interactions — which scatter into every basis state — repopulate them. Why: you cannot consistently truncate to positive energies, so the theory itself insists you interpret the negative branch, and the interpretation is the antiparticle. *In summary: the Klein–Gordon density is indefinite, while the Dirac density $\psi^\dagger\psi\ge0$ is positive definite; both equations share the spectrum $E=\pm\sqrt{p^2+m^2}$ , and there are four spinor solutions per momentum — two at each energy sign.*

Go deeper · Derivation B · 1 minCharge conjugation turns an electron solution into a positron

Derivation BCharge conjugation turns an electron solution into a positron

The aim is to show, at the level of the equation rather than by assertion, that the negative-energy partner is a particle of the same mass and opposite charge — that "antiparticle" is a derived fact, not a tacked-on definition. We couple the Dirac field to electromagnetism, apply charge conjugation to the whole equation, and read off what field the new solution responds to — the operational meaning of $C$ .

Step 1. Minimally couple: the electron of charge $-e$ obeys $\big[i\gamma^\mu(\partial_\mu+ieA_\mu)-m\big]\psi=0$ , the $+ieA_\mu$ encoding its coupling to the field. Why: the sign of the coupling term is precisely what distinguishes how a positive versus negative charge moves in a given field, so it is the quantity we must track.

Step 2. Take the complex conjugate and left-multiply by $i\gamma^2$ . Using the Dirac-basis identity $\gamma^2\gamma^{\mu*}\gamma^2=-\gamma^\mu$ , the derivative and mass terms keep their form while the coupling flips sign; defining $\psi^c=i\gamma^2\psi^*$ gives $\big[i\gamma^\mu(\partial_\mu-ieA_\mu)-m\big]\psi^c=0$ . Why: this is an identity, not an approximation — the same field $A_\mu$ now couples with the opposite sign, exactly as a charge $+e$ would.

Step 3. Read the result: $\psi^c$ satisfies the Dirac equation for a particle of the same mass $m$ but charge $+e$ . Applied to a negative-energy, spin-down electron solution, $C$ delivers a positive-energy, spin-up positron solution. Why: mass is untouched and charge is reversed — the two defining properties of the antiparticle drop straight out of the algebra.

Step 4. Iterating, $C$ reverses every additive charge (electric charge, baryon number, lepton number, strangeness) while leaving mass, spin, and lifetime invariant. A particle with no additive charges can be its own antiparticle — the photon, and possibly the neutrino. Why: this classifies the whole zoo — charged fermions come in distinct pairs, truly neutral bosons are self-conjugate — from one operation. *In summary: $\psi^c=i\gamma^2\psi^*$ obeys the opposite-charge Dirac equation with unchanged mass $m$ , and $C$ flips additive quantum numbers only.*

Go deeper · Derivation C · 1 minHow CPT forces equal masses and lifetimes

Derivation CHow CPT forces equal masses and lifetimes

The goal is to sketch why the equality of particle and antiparticle mass is not an empirical coincidence but a theorem — following from three assumptions so basic that abandoning any of them would unravel relativistic quantum theory. We identify the three ingredients, state what the combined operation does to a one-particle state, and extract the mass and lifetime equalities from Hermiticity.

Step 1. The three assumptions are Lorentz invariance, locality (fields at spacelike separation commute or anticommute — microcausality), and a Hermitian Hamiltonian bounded below; spin-statistics follows from the same premises. Why: these are exactly the axioms of any sensible relativistic QFT, so the theorem's conclusions are as secure as the framework itself.

Step 2. The antiunitary operator $\Theta=CPT$ maps a particle state $|a;\mathbf p,s\rangle$ to the corresponding antiparticle state $|\bar a;\mathbf p,-s\rangle$ (reversing charge via $C$ , momentum via $P$ , and spin/motion via $T$ ). The Lüders–Pauli theorem proves $[\Theta,H]=0$ for any theory obeying Step 1. Why: this is the content of the theorem — the combined reflection is an exact symmetry even though $C$ , $P$ , $T$ individually are not.

Step 3. A stable particle's mass is the eigenvalue $H|a;\mathbf 0\rangle=m_a|a;\mathbf 0\rangle$ . Apply $\Theta$ : since $\Theta H=H\Theta$ , the antiparticle state $\Theta|a;\mathbf 0\rangle=|\bar a;\mathbf 0\rangle$ satisfies $H|\bar a;\mathbf 0\rangle=m_a|\bar a;\mathbf 0\rangle$ , so $m_{\bar a}=m_a$ . Why: the mass is a Hamiltonian eigenvalue, and a symmetry of $H$ maps eigenstates to eigenstates of the same eigenvalue — equal masses are immediate.

Step 4. For an unstable particle the same argument applied to the total decay width $\Gamma=\hbar/\tau$ (fixed by a Hermitian interaction summed over all channels) gives $\Gamma_{\bar a}=\Gamma_a$ , hence $\tau_{\bar a}=\tau_a$ — even though individual partial widths and CP asymmetries may differ. Why: CPT equalizes the total lifetime exactly while permitting the channel-by-channel CP violation that lets the universe tell matter from antimatter. *In summary: $[\Theta,H]=0$ under Lorentz invariance, locality, and Hermiticity implies $m_a=m_{\bar a}$ and total $\Gamma_a=\Gamma_{\bar a}$ ; partial widths need not match.*

Limiting cases

Case	Result	Meaning
Charged fermion (e⁻)	distinct antiparticle $e^+$	same mass $m_e$ , opposite charge; CPT pairs them exactly
Self-conjugate boson (γ)	particle = antiparticle	photon carries no additive charge, so $C\gamma=-\gamma$ up to sign; it is its own antiparticle
Truly neutral fermion (ν?)	Majorana: $\nu=\bar\nu$ possible	allowed only if all additive charges vanish; open experimental question for neutrinos
Neutral but flavoured ( $K^0$ )	$\bar K^0\ne K^0$	distinct because strangeness $S=+1$ vs $-1$ ; their mixing yields CP violation
Non-relativistic limit	antiparticles decouple	pair creation costs $2mc^2\gg$ kinetic energies, sea inert; ordinary one-particle QM recovered

Common trapAntiparticles are just particles running backward in time

This is a calculational reinterpretation, not a literal fact about the world. In the Feynman–Stückelberg trick a negative-energy solution propagating backward is bookkept as a positive-energy antiparticle propagating forward — compact for diagrams, and it correctly gives that emitting an antiparticle of momentum $p$ equals absorbing a particle of momentum $-p$ . But a real positron in a detector is an entirely ordinary, positive-energy particle moving forward in time; nothing travels to the past. Two further traps: an antiparticle is not "a particle with a minus sign" — it has the same positive mass and energy, only its additive quantum numbers reverse. And CPT does not make matter and antimatter interchangeable in every process: $C$ , $P$ , and $T$ are each violated, and it is the surviving CP violation that distinguishes the two and underlies the matter dominance of the universe.

Worked examples

1 · Pair creation and annihilation of the electron (fully worked)

A photon converts to an electron–positron pair in the Coulomb field of a nucleus. Find the threshold photon energy. Then take an electron and positron annihilating from rest and find the energy and number of photons produced.

Step 1. Creating the pair costs at least the two rest energies. With $m_e c^2=0.511$ MeV, the minimum is $2m_ec^2=2(0.511)=\mathbf{1.022\ MeV}$ . (A free photon in vacuum cannot conserve four-momentum while doing this — a nearby nucleus absorbs the recoil — but the energy threshold is set by the rest masses.)

Step 2. Run it backward: an $e^+e^-$ pair at rest has total energy $2m_ec^2=1.022$ MeV and zero net momentum. It cannot go to a single photon, because a photon must carry momentum $E/c\ne0$ — momentum conservation forbids it. The minimum is two photons, emitted back-to-back so their momenta cancel.

Step 3. By energy conservation the two photons share $1.022$ MeV equally, so each has $E_\gamma=\mathbf{0.511\ MeV}$ , back to back. This is the signal PET scanners detect: the two 511 keV photons fire a coincidence line through the annihilation point.

Step 4. Ortho-positronium (spin-1, $C=-1$ ) cannot decay to two photons — a two-photon state has $C=(-1)^2=+1$ — so it decays to three photons with a continuous energy spectrum, a direct experimental check of the charge-conjugation selection rule $C_\gamma^{\,n}=(-1)^n$ . Threshold 1.022 MeV; annihilation → 2γ at 0.511 MeV each (para) or 3γ (ortho).

2 · The antiproton production threshold (fully worked)

Antiprotons were first made by firing protons at a fixed hydrogen target via $p+p\to p+p+p+\bar p$ . Baryon-number conservation forbids making a lone antiproton, so a proton–antiproton pair appears alongside the two original protons. Find the minimum kinetic energy of the beam proton.

Step 1. The lightest allowed final state is four baryons — $p,p,p,\bar p$ — at rest in their common centre-of-mass frame, so the minimum invariant mass is $\sqrt{s_{\min}}=4m_p$ . Using the invariant $s$ is the clean way to handle the fixed-target kinematics.

Step 2. For a beam proton of total energy $E$ hitting a proton at rest, $s=2m_p^2+2m_pE$ . Setting this to $s_{\min}=(4m_p)^2=16m_p^2$ gives $2m_p^2+2m_pE=16m_p^2$ , so $E=7m_p$ .

Step 3. The beam kinetic energy is $T=E-m_pc^2=7m_pc^2-m_pc^2=\mathbf{6\,m_pc^2}$ . With $m_pc^2=0.938$ GeV, $T=\mathbf{5.63\ GeV}$ .

Step 4. This is why the Berkeley Bevatron was designed for $6.2$ GeV — comfortably above threshold — and in 1955 Chamberlain and Segrè used it to discover the antiproton, confirming the proton too has its CPT partner. Threshold $T=6m_pc^2\approx5.63$ GeV, matching the 6.2 GeV Bevatron.

3 · How equal is "equal mass"?

CPT predicts $m_{e^+}=m_{e^-}$ exactly. How is this tested, and how tightly?

Step 1. In a Penning trap a single lepton's cyclotron and spin-precession frequencies fix its $g$ -factor. Comparing electron and positron in the same trap, Dehmelt's group found the anomalies agree to $(g_{e^-}-g_{e^+})/g_{\text{avg}}=(-0.5\pm2.1)\times10^{-12}$ — zero at the part-per-trillion level.

Step 2. The neutral kaons do even better on mass: because $K^0$ and $\bar K^0$ mix, their tiny mass split is amplified into an observable oscillation, bounding $|m_{K^0}-m_{\bar K^0}|/m_{K^0}<10^{-18}$ — the most precise mass comparison in physics. CPT survives at $10^{-12}$ for the electron and below $10^{-18}$ for the kaon; no violation has ever been seen.

Physics of the portrait

Antimatter is one of those rare ideas that a single person can be said to have created from pure theory. It carries Dirac's fingerprints from the equation that predicted it to the reluctant physical picture he built to explain the states he could not delete.

Origins · 1 minPaul Dirac and how this was discovered

Paul Dirac

1902–1984 · relativistic quantum mechanics and the prediction of antimatter

Paul Adrien Maurice Dirac was the most austere of the quantum pioneers — a Cambridge engineer-turned-physicist who trusted mathematical beauty over physical intuition and spoke almost never. In 1928, seeking a wave equation for the electron consistent with special relativity and carrying a positive probability density, he factorized the relativistic energy relation and arrived at the four-component spinor equation that bears his name. It handed him, as a bonus, the electron's spin and magnetic moment $g=2$ — but also solutions of negative energy he could not throw away. His first fix was the hole theory: an infinite filled sea of negative-energy electrons whose vacancies appear as positive charges. He first guessed the holes were protons, but Weyl and Oppenheimer showed a hole must have the electron's mass, so in 1931 Dirac predicted a brand-new particle — "a new kind of particle, unknown to experimental physics, having the same mass and opposite charge to an electron", the anti-electron. Carl Anderson found it in cosmic rays the following year and named it the positron; Dirac shared the 1933 Nobel Prize with Schrödinger. Hole theory was later superseded by quantum field theory and the Feynman–Stückelberg picture, and antimatter was welded to matter by the CPT theorem of Lüders and Pauli — but the prediction stands as perhaps the purest case of new physics deduced from the consistency of an equation.

History

1928

— Dirac writes the relativistic wave equation for the electron

1931

— Dirac predicts an anti-electron of the same mass and opposite charge

1932

— Anderson photographs the positron in cosmic-ray cloud-chamber tracks

1934

— Pauli and Weisskopf show charged bosons also have antiparticles, no sea required

1955

— Chamberlain and Segrè produce the antiproton at the Berkeley Bevatron; Lüders and Pauli complete the CPT theorem

Going further

Texts. Peskin & Schroeder, An Introduction to Quantum Field Theory, §3.3–3.5 (Dirac field, quantization, discrete symmetries) · Griffiths, Introduction to Elementary Particles, 2nd ed., Ch. 1 and §7.3 (charge conjugation) · Halzen & Martin, Quarks and Leptons, Ch. 3–5 (Dirac equation, antiparticles, Feynman–Stückelberg) · Thomson, Modern Particle Physics, Ch. 4 (the Dirac equation and antiparticles) · Schwartz, Quantum Field Theory and the Standard Model, Ch. 11 (the spin-statistics and CPT theorems). The key results collected in this article are: both energy signs survive the Dirac linearization while the density becomes positive definite; $\psi^c=i\gamma^2\psi^*$ obeys the opposite-charge Dirac equation with unchanged mass; the CPT consequences $m_a=m_{\bar a}$ , $\tau_a=\tau_{\bar a}$ ; and the numeric anchors $2m_ec^2=1.022$ MeV, annihilation photons $0.511$ MeV, antiproton threshold $T=6m_pc^2=5.63$ GeV. The experimental bounds (kaon $|\Delta m|/m<10^{-18}$ , $e^\pm$ $g$ -factor $\sim10^{-12}$ , BASE 16 ppt, ALPHA $1S$ – $2S$ $2\times10^{-12}$ ) are current measured values.

Prerequisites. The Dirac Equation · Special-Relativistic Kinematics · Standard Model Overview. Leads to. Dirac Solutions and Spinors · Hole Theory and the Dirac Sea · Matter–Antimatter Asymmetry.

Ask the tutor: Ask me why the Dirac equation cannot escape its negative-energy solutions, or to work the antiproton threshold or a CPT bound step by step.

Is an antiparticle just a particle with the charges flipped, or is it a genuinely different object?

Both, correctly stated. An antiparticle has exactly the same mass, spin, and lifetime as its partner — fixed equal by the CPT theorem — and every one of its additive quantum numbers (electric charge, baryon number, lepton number, strangeness) is reversed. So the positron is not "a lighter or negative-energy electron"; it is a full positive-energy particle carrying the opposite sign of every additive charge. For a particle whose additive charges all vanish there is nothing left to flip, and it can be its own antiparticle — the photon, and possibly the neutrino.

Did Dirac really predict antimatter, or did he just tolerate strange solutions?

He predicted it, though not on the first try. The negative-energy solutions were there in the 1928 equation, and at first Dirac hoped the holes in his filled sea might be protons — the only other particle then known. Weyl and Oppenheimer proved a hole must have exactly the electron's mass, not the proton's, leaving Dirac a stark choice: abandon the theory or predict a new particle. In 1931 he committed to "an anti-electron", a particle of electronic mass and positive charge, and Anderson's 1932 discovery of the positron vindicated him. It remains one of physics' cleanest cases of a particle deduced from theory before any experiment demanded it.

If antiparticles are "negative-energy states running backward in time", how can a real positron have positive energy?

Because the "backward in time" language is a bookkeeping device, not a claim about a real trajectory. In the Feynman–Stückelberg scheme a mathematically negative-energy solution moving backward is reinterpreted as a physical positive-energy antiparticle moving forward — the two give identical amplitudes, and the second is the physical one. A positron in your lab is an utterly ordinary particle: positive rest energy $0.511$ MeV, moving forward in time, leaving a track that curves opposite to an electron's in a magnetic field. Nothing travels to the past; the arrow-reversal lives only inside the calculation, where it lets one propagator handle particles and antiparticles at once.

What exactly is the CPT theorem, and why is it considered so unbreakable?

CPT says that the combined operation of charge conjugation (C), parity (P), and time reversal (T) is an exact symmetry of any quantum field theory that is Lorentz invariant, local (fields at spacelike separation commute or anticommute), and has a Hermitian Hamiltonian. Proved by Lüders, Pauli, and Bell in the mid-1950s, it forces particle and antiparticle to share mass and total lifetime and to have exactly opposite charge and magnetic moment. It is considered near-inviolable because breaking it would require abandoning relativity, locality, or unitarity — pillars whose failure would signal a revolution. Crucially, CPT does not require C, P, or T to hold individually, and all three are separately violated in nature.

If CPT makes matter and antimatter so symmetric, why is the universe made of matter?

This is one of the great open problems, and CPT is exactly why it is hard. CPT guarantees equal masses and equal total decay rates, so the early universe should have made matter and antimatter in equal amounts and left nothing but radiation after they annihilated. The escape is that CPT constrains only total rates; it permits CP violation, small differences between specific matter and antimatter processes. Sakharov showed in 1967 that CP violation, together with baryon-number violation and departure from thermal equilibrium, could generate a matter excess. The Standard Model's CP violation (in the CKM matrix) is far too small to account for the observed asymmetry, so an additional source — new physics — must exist.

Do antiparticles feel gravity the same way, or could there be antigravity?

They feel it the same way, and this has now been measured rather than assumed. Because a particle and its antiparticle have equal mass by CPT, the equivalence principle predicts they fall identically in a gravitational field. Testing it directly is hard — antimatter annihilates on contact with ordinary matter — but in 2023 the ALPHA-g experiment at CERN released cold antihydrogen atoms in a vertical trap and watched them fall downward under Earth's gravity, with an acceleration consistent with $g$ and firmly excluding "antigravity". Combined with the agreement of antihydrogen's spectrum with hydrogen's, this closes a long-standing speculation: antimatter is gravitationally ordinary matter with the charges reversed.