56
Part Two · Quantum fields and particles

The Klein–Gordon equation

In the winter of 1925, before he found the equation that bears his name, Schrödinger wrote down a relativistic one. He applied it to hydrogen, got the fine structure wrong, and put it aside. Oskar Klein and Walter Gordon published it in 1926, and for a decade it seemed a dead end: its probability could go negative. It came back as the equation of every particle with no spin. The Higgs boson obeys it.

On the plate: a Klein–Gordon wave packet made of positive- and negative-frequency parts.

The equation

Take E2=p2+m2E^2 = \mathbf p^2 + m^2 and replace E→i∂tE \to i\partial_t, p→−i∇\mathbf p \to -i\nabla, as in Chapter 31:

(∂t2−∇2+m2)ϕ=0⟺(∂μ∂μ+m2) ϕ=0\left(\partial_t^2 - \nabla^2 + m^2\right)\phi = 0 \quad\Longleftrightarrow\quad (\partial_\mu\partial^\mu + m^2)\,\phi = 0

It is second order in time and manifestly relativistic. Its plane waves e−i(Et−p⋅x)e^{-i(Et - \mathbf p\cdot\mathbf x)} come with both signs of EE. A packet travels at the group velocity p/Ep/E, always below light speed. Established

A density that can go negative

Like the Schrödinger equation, Klein–Gordon has a conserved current, jμ=i(ϕ∗∂μϕ−ϕ ∂μϕ∗)j^\mu = i(\phi^*\partial^\mu\phi - \phi\,\partial^\mu\phi^*), with ∂μjμ=0\partial_\mu j^\mu = 0. But its time component, ρ=i(ϕ∗ϕ˙−ϕϕ˙∗)\rho = i(\phi^*\dot\phi - \phi\dot\phi^*), is not ∣ϕ∣2|\phi|^2. For a positive-frequency wave it is positive; for a negative-frequency wave, negative.

Mix in negative frequencies: 0.30. The plate shows ρ\rho: amber where positive, red where negative. The two parts move apart, and ρ\rho cannot be a probability. In 1934 Wolfgang Pauli and Victor Weisskopf reinterpreted it as charge density: the positive parts are particles, the negative parts antiparticles of opposite charge.

Negative energy, read correctly

A negative-frequency solution moving forward in time, e+iEte^{+iEt}, has the same form as a positive-energy one moving backward. Ernst Stückelberg and later Feynman read it that way: a negative-energy particle going into the past is an antiparticle with positive energy going into the future. Nothing ever has negative energy. In the field theory of Chapter 60, the negative-frequency part of ϕ\phi becomes the operator that creates antiparticles, and the energy of every state is positive.

Forces from massive particles

Put a point source in the static equation: (−∇2+m2)ϕ=g δ3(r)(-\nabla^2 + m^2)\phi = g\,\delta^3(\mathbf r). The solution is

ϕ(r)=g4π e−mrr\phi(r) = \frac{g}{4\pi}\,\frac{e^{-mr}}{r}

Coulomb’s 1/r1/r, cut off beyond a distance 1/m1/m. Hideki Yukawa proposed in 1935 that the force holding nuclei together comes from exchanging a particle of mass about 100–200 MeV, to give a range of about 1 fm. The pion, found in 1947 at 139.6 MeV, fits: 1/m=1.411/m = 1.41 fm. Set the mediator mass to 140 MeV; zero mass gives back Coulomb’s law. Heavy mediators give short-range forces: the W and Z, at 80–91 GeV, reach only about 0.002 fm, which is why the weak force is weak. Established

Where it lives now

As a single-particle wave equation, Klein–Gordon fails, for the reasons of Chapter 55. As a field equation it is exactly right for spin-0 particles: pions in effective theories, and the Higgs field, whose excitation was discovered in 2012 at 125 GeV. Schrödinger’s discarded equation turned out to be the first piece of the Standard Model.

What you now have

The Klein–Gordon equation, its conserved current and why its density can be negative, the charge reinterpretation, antiparticles as negative frequencies, and the Yukawa potential with its range 1/m1/m.

· · ·

Three questions to carry with you.

Why isn’t ρ a probability density?tap
It is negative for negative-frequency solutions; it is a charge density instead.
What sets the range of a Yukawa force?tap
The mediator’s mass: range ≈ 1/m = ħ/mc.
Which known elementary particle obeys Klein–Gordon?tap
The Higgs boson, the only elementary spin-0 particle found so far.

And one to work out. The pion has mass 139.57 MeV. What is the range of the Yukawa force it carries, ℏc/m\hbar c/m, in fm?

range =fm

Sources: O. Klein, Z. Phys. 37, 895 (1926). W. Gordon, Z. Phys. 40, 117 (1926). W. Pauli and V. Weisskopf, Helv. Phys. Acta 7, 709 (1934). H. Yukawa, Proc. Phys.-Math. Soc. Japan 17, 48 (1935). M. D. Schwartz, Quantum Field Theory and the Standard Model. Checked by code: ranges 1.414 fm (π), 0.0022 fm (Z), 0.0016 fm (Higgs); KG packet density sign follows frequency sign.

Next · Chapter 57
The Dirac equation and antimatter

Dirac wanted an equation linear in time. Out came spin, g = 2, and a prediction of antimatter.

Turn the page