59
Part Two · Quantum fields and particles

Noether's theorem

Why is energy conserved? Not because someone measured it carefully, though they did. Energy is conserved because the laws of physics are the same today as they will be tomorrow. Momentum is conserved because they are the same here as over there; angular momentum, because they don’t care which way you face; electric charge, because the phase of a quantum field can be turned without anyone noticing. In 1918 Emmy Noether proved that every continuous symmetry of this kind comes with a conserved quantity, and that the connection runs both ways.

On the plate: a particle orbiting in a perfectly round bowl. Its angular momentum never changes.

Break the symmetry, lose the conservation

A particle moves in the bowl V=12[(1+ε)x2+y2]V = \tfrac12[(1 + \varepsilon)x^2 + y^2]. When ε=0\varepsilon = 0 the bowl is round, the Lagrangian is unchanged by rotations, and the angular momentum LzL_z is constant. Squash the bowl with ε=\varepsilon = 0.20. Now LzL_z, plotted in amber, swings back and forth, even reversing sign. The energy, in white, stays perfectly flat: the bowl still does not change with time.

This is Noether’s theorem in miniature. A conserved quantity is not a separate law; it is the shadow of a symmetry. Established

The theorem

Suppose a small change of the field, ϕ→ϕ+ϵ δϕ\phi \to \phi + \epsilon\,\delta\phi, changes the Lagrangian density at most by a total derivative, δL=ϵ ∂μKμ\delta\mathcal L = \epsilon\,\partial_\mu K^\mu. Then, on any solution of the equations of motion,

jμ=∂L∂(∂μϕ) δϕ−Kμobeys∂μjμ=0j^\mu = \frac{\partial\mathcal L}{\partial(\partial_\mu\phi)}\,\delta\phi - K^\mu \qquad\text{obeys}\qquad \partial_\mu j^\mu = 0

A current with zero divergence means the charge Q=∫j0 d3xQ = \int j^0\,d^3x in any region changes only by what flows through its boundary. The plate shows it: the amount in the box changes only as current crosses the edge, never by appearing or vanishing inside.

DeeperCompute δL\delta\mathcal L by the chain rule, use the Euler–Lagrange equation of Chapter 58 to replace ∂L/∂ϕ\partial\mathcal L/\partial\phi, and the result collects into ∂μ(…)\partial_\mu(\ldots). Setting it equal to ∂μKμ\partial_\mu K^\mu gives the theorem.

Energy, momentum and the stress tensor

Shifting all of spacetime, xν→xν+aνx^\nu \to x^\nu + a^\nu, is four symmetries at once, and they give four conserved currents, gathered into the stress–energy tensor:

Tμν=∂L∂(∂μϕ) ∂νϕ−ημνL,∂μTμν=0T^{\mu\nu} = \frac{\partial\mathcal L}{\partial(\partial_\mu\phi)}\,\partial^\nu\phi - \eta^{\mu\nu}\mathcal L, \qquad \partial_\mu T^{\mu\nu} = 0

T00T^{00} is the energy density, T0iT^{0i} the momentum density, and TijT^{ij} the flow of momentum, which is pressure and stress. Choose or to see its pattern. In general relativity this same tensor is what curves spacetime.

Phase symmetry and charge

A complex field can have its phase turned everywhere at once, ϕ→eiαϕ\phi \to e^{i\alpha}\phi, and the Lagrangian ∣∂ϕ∣2−m2∣ϕ∣2|\partial\phi|^2 - m^2|\phi|^2 does not notice. Noether gives the current of Chapter 56, jμ=i(ϕ∗∂μϕ−ϕ ∂μϕ∗)j^\mu = i(\phi^*\partial^\mu\phi - \phi\,\partial^\mu\phi^*); for the Dirac field it is ψˉγμψ\bar\psi\gamma^\mu\psi. Conservation of electric charge is this symmetry. In Chapter 74, demanding that the phase may be turned differently at each point will force the photon to exist.

When energy is not conserved

The theorem works both ways: no symmetry, no guarantee. An expanding universe is not the same today as tomorrow, and light crossing it loses energy. The cosmic microwave background was released when the universe was 1,100 times smaller; each photon then carried about 0.26 eV and now carries about a thousandth of that. Where did the energy go? Nowhere: in an expanding spacetime there is in general no conserved total energy to account for. Energy is still conserved locally, ∇μTμν=0\nabla_\mu T^{\mu\nu} = 0, but a global total needs a time symmetry that the universe does not have. Established, though how to phrase it is a matter of taste.

What you now have

Noether’s theorem and its proof, conserved currents and charges, the stress–energy tensor, phase symmetry and charge conservation, and why energy is not globally conserved in an expanding universe.

· · ·

Three questions to carry with you.

Which symmetry gives energy conservation?tap
Invariance under shifts in time.
Which symmetry gives electric-charge conservation?tap
Invariance under a global change of phase of the charged fields.
Why does light lose energy crossing an expanding universe?tap
The expanding spacetime has no time-translation symmetry, so there is no conserved global energy.

And one to work out. A CMB photon had 0.26 eV when the universe was 1,101 times smaller than now. What is its energy today, in meV?

E =meV

Sources: E. Noether, Nachr. Ges. Wiss. Göttingen, 235 (1918); English translation by M. A. Tavel, Transport Theory Stat. Phys. 1, 186 (1971). M. D. Schwartz, Quantum Field Theory and the Standard Model, ch. 3. S. M. Carroll, Spacetime and Geometry, ch. 8. Checked by code: L_z constant for ε = 0; for ε = 0.2, L_z runs 0.70 → −0.74 over t = 30 while energy stays 0.845 to six figures; CMB photon 0.236 meV.

Next · Chapter 60
Canonical quantization of the scalar field

Quantize a field and particles appear as its excitations. Particles are not put in by hand.

Turn the page