running of the strong coupling
We usually think of the strength of a force as a fixed number. But in the quantum world, the apparent strength of an interaction depends on how closely you look — on the energy of the probe you use. Physicists call this changing strength a "running" coupling, and the strong force has the most dramatic running of all. The number that measures the strong force's strength, written alpha_s, is large at low energies and shrinks as the energy rises.
Why does the strength change at all? Empty space near a charge is not really empty: it seethes with short-lived virtual particles that partly screen or anti-screen the charge. For electromagnetism, this cloud screens the charge, so the force looks a little stronger the closer you probe. For the strong force, the gluons' self-interaction wins out and does the opposite: it "anti-screens," so the force looks weaker the closer (the higher the energy) you probe. That downward trend in alpha_s with energy is exactly asymptotic freedom, expressed as a number.
The running is not a hand-wave — it is measured. At an energy around the mass of the Z boson (about 91 GeV) alpha_s is roughly 0.118; at lower energies it climbs toward 1 and beyond, which is where calculations stop being reliable and confinement takes over. Experiments at many different energies trace out the falling curve, and it matches QCD's prediction precisely — one of the sharpest confirmations that the theory is right. The mathematical tool that describes how couplings run with energy is called the renormalization group.
Measured values of alpha_s fall from roughly 0.3 at a few GeV to about 0.118 at 91 GeV, exactly tracing QCD's predicted curve.
The strong coupling alpha_s shrinks as the probing energy rises — measured across decades of energy.
Calling alpha_s a "constant" is a historical misnomer; it is not constant but varies with energy. The same is true, more mildly, of the electromagnetic coupling. There is no single "strength of the strong force" — only a strength at a stated scale.