78
Part Two · Quantum fields and particles

Quantum chromodynamics

The Δ⁺⁺ baryon is three up quarks with all their spins pointing the same way, in the same lowest-energy state. That should be impossible: quarks are fermions, and the exclusion principle (Chapter 43) forbids three identical fermions in one state. The fix, proposed in 1964 and 1965, was that quarks carry a hidden label with three values, now called colour, and the three quarks in a Δ⁺⁺ each have a different one. A decade later, counting the rate at which electron–positron collisions make hadrons confirmed the number three. Colour is the charge of the strong force, and its theory is quantum chromodynamics.

On the plate: the hadron-production ratio R against energy, stepping up at each new quark.

The Lagrangian

QCD is Yang–Mills theory (Chapter 75) with gauge group SU(3) acting on colour, plus quarks:

LQCD=∑qqˉ (iγμDμ−mq) q−14GμνaGaμν\mathcal L_{\text{QCD}} = \sum_q\bar q\,(i\gamma^\mu D_\mu - m_q)\,q - \tfrac14G^a_{\mu\nu}G^{a\mu\nu}

Each quark field is a colour triplet, red, green, blue; there are six flavours, u, d, s, c, b, t. SU(3) has 32−1=83^2 - 1 = 8 generators, so there are eight gluons, each carrying a colour and an anticolour. Not nine: the colourless combination rrˉ+ggˉ+bbˉr\bar r + g\bar g + b\bar b is the singlet, not part of the octet, so there is no long-range gluon force between colourless hadrons. Harald Fritzsch, Murray Gell-Mann and Heinrich Leutwyler proposed the theory in 1973. Established

Counting colours

At high energy, e+e−e^+e^- annihilates into a quark–antiquark pair just as into a muon pair (Chapter 67), but with the quark’s charge QqQ_q and once for each colour. So the ratio

R=σ(e+e−→hadrons)σ(e+e−→μ+μ−)=Nc∑qQq2R = \frac{\sigma(e^+e^- \to \text{hadrons})}{\sigma(e^+e^- \to \mu^+\mu^-)} = N_c\sum_q Q_q^2

counts colours directly. With u, d, s: 3×(49+19+19)=23\times(\tfrac49 + \tfrac19 + \tfrac19) = 2. Above charm threshold, 10/310/3; above bottom, 11/311/3. With Nc=1N_c = 1, the steps would be a third as high. Choose or . Measurements from 1974 on, collected by the Particle Data Group, follow the three-colour steps, a few percent higher because of the gluon-emission correction 1+αs/π1 + \alpha_s/\pi. The spikes are bound states: the J/ψ at 3.10 GeV and the Υ at 9.46 GeV. Established

Colour factors

How strongly a quark emits a gluon is set by the Casimir of its representation (Chapter 73), CF=4/3C_F = 4/3; how strongly a gluon emits a gluon, by CA=3C_A = 3. Gluons radiate more than twice as readily as quarks, which is why a jet started by a gluon is broader and contains more particles than one started by a quark, a difference measured at LEP and the LHC and used to tell the two apart.

Only colourless things are free

Every hadron ever seen is a colour singlet: three quarks of different colours (baryons) or a quark and antiquark of colour and anticolour (mesons). Individual quarks and gluons have never been isolated. QCD predicts this, though the proof is still missing, as the next chapter explains.

What you now have

Why colour was needed, the QCD Lagrangian, eight gluons, the R ratio and the measurement of three colours, colour factors CFC_F and CAC_A, and colour singlets.

· · ·

Three questions to carry with you.

Why are there 8 gluons and not 9?tap
SU(3) has 8 generators; the colour-singlet combination is not part of the gauge group.
What does R measure?tap
N_c times the sum of squared charges of quarks light enough to be produced.
Why do gluon jets look broader?tap
Gluons radiate with colour factor C_A = 3, quarks with C_F = 4/3.

And one to work out. What is R (without the QCD correction) above the bottom threshold, with u, d, s, c, b?

R =

Sources: O. W. Greenberg, Phys. Rev. Lett. 13, 598 (1964); M. Y. Han and Y. Nambu, Phys. Rev. 139, B1006 (1965). H. Fritzsch, M. Gell-Mann and H. Leutwyler, Phys. Lett. B 47, 365 (1973). Particle Data Group, “Plots of cross sections and related quantities” (R ratio compilation). Checked by code: R = 2, 10/3, 11/3; with α_s/π correction ≈ 2.13, 3.55, 3.90.

Next · Chapter 79
Asymptotic freedom and confinement

Quarks are free when close together and locked in forever when pulled apart. Proving the second part is worth $1 million.

Turn the page