The Confinement

She held out her hand, palm open, and asked for a quark.

The physicist smiled. "You can't. Quarks don't come out of hadrons. You can pull them apart and give them enough energy, and still they won't come out. They'll just make more hadrons."

"Like a rubber band?"

"Like a rubber band that gets stronger the more you stretch it. It's the color field. When it snaps, it produces new quark-antiquark pairs from the vacuum. You end up with two mesons, not two quarks."

This is color confinement. One of the Clay Mathematics Institute's Millennium Prize Problems. Despite fifty years of studying QCD, we do not have a rigorous proof. We believe it. We see it. We simulate it.

Here is what we know, from the lattice.

Put a static quark and antiquark a distance r apart. Compute the energy. It grows linearly:

V(r) ≈ σr

where σ is the string tension, approximately (440 MeV)². The force is constant. Not 1/r². Constant. Like a rubber band.

Pull them apart, and the potential energy grows without bound. When it exceeds 2m_q, the string snaps — not into two loose quarks, but into two mesons. The flux tube breaks. The newly created quark pairs with the original antiquark. Color neutrality at every step. You can never isolate color.

In QCD, pulling a quark injects energy into a flux tube storing energy linearly. There is no ionization energy. Only the string tension.

The flux tube is real. On the lattice, you measure gluon field energy density between a static pair. It concentrates into a tube about 0.5 fm radius. In heavy mesons like ψ and Υ, the tube is clean. In light mesons, messier, but present.

String breaking appeared on the lattice in 2008. The potential flattened at large separation. But even at infinite distance, the states are color-neutral. Always.

Trolla thinks about confinement in terms of topology. The QCD vacuum is a medium, and the flux tube is a topological defect — a region where the order parameter differs from its ground state. The confinement phase is unbroken. The deconfinement transition at T_c is a phase transition where the vacuum reorganizes and color charges can roam. We see this in the early universe and in heavy-ion collisions. Confinement is the ground state. Deconfinement is the excited state.

The unproved part is showing this follows from the Yang-Mills Lagrangian alone. Lattice simulations show it. Perturbation theory cannot. The Wilson area law is the diagnostic, but deriving it from the continuum path integral eludes us.

Confinement remains: a fact of nature, a feature of the theory, an open question in mathematics.