It is often said that we could probe quark gluon plasma with photon or dileptons, because they interact electromagnetically so tha tthey don't modify the medium.

But do they do not modify the medium ?

It looks to me that the photon could create quarks for example. Same for dileptons : they could produce photons, that produce quarks.

So why do photon and dilepton not interacting in quark gluon plasma ?


2 Answers 2


The strong force is by far the dominant interaction in quark-gluon plasma. The strong coupling constant is simply much bigger than the electromagnetic one, even at that energy. So photons and dileptons are said to be penetrating probes of heavy-ion collisions because they're color-neutral.

That said, photons and leptons still do occasionally interact with quark-gluon plasma, and studies of photon and dilepton probes have to account for this. But, due to their color neutrality, their mean free path in dense nuclear matter is much longer than quarks or gluons, so most photons and dileptons escape mostly unaltered.

For a more detailed review of the above concepts, see e.g. https://arxiv.org/pdf/1907.08893.pdf. In this review, you will find an estimate of the mean free path for photons in QGP; for $\alpha_s=0.4$ and $T=200$ MeV, the mean free path of a 2-GeV photon in QGP is nearly $500$ fm! This is many times larger than the size of the QGP droplet, which is on the order of 10 fm wide.

  • $\begingroup$ Note that 500 fm really is (!)-big. A uranium nucleus is only 15 fm, and that's about as big as any scattering target gets. $\endgroup$
    – JEB
    Commented Apr 28, 2020 at 23:08
  • $\begingroup$ Thank you @probably_someone $\endgroup$ Commented Apr 29, 2020 at 7:05

Vector meson dominance (VMD) was a pre-QCD treatment of the hadronic content of the photon in which the propagator went something like this:

enter image description here

My understanding is that it is applicative to soft processes.

A similar treatment appears in polarized semi inclusive deep inelastic scattering (DIS) experiments:

$$\vec e(\vec N, e'\pi^+\pi^-)N $$ $$\vec e(\vec N, e'K^+K^-)N $$

Here the $\vec e$ and $e'$ serve to define a (polarized) virtual photon, and the detected pions (kaons) reconstruct to a rho (phi) meson that is the photon that has coherently scattered from the polarized nucleon (or nucleus). The invariant mass of the final state may also indicate a resonance, but again: that's a coherent process, not hard scattering off point partons.

(What makes this reaction interesting is that the dependence on the decay plane angle defined by the pseudo scalar mesons relative to the electron scattering plane is sensitive to the vector meson polarization....all while the virtual photon polarization is known).


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.