Light-by-light scattering proves to be extremely delicate to observe, as the probability for two photos to interact in this manner is very weak. For this reason, one single scattering requires very high photon fluxes.

The collisions of lead ions at the LHC, however, provide the ideal experimental conditions to study this interaction. When lead ions are accelerated at ultra-relativistic speeds, their intense electromagnetic fields create the equivalent of a tremendous flow of photons surrounding them. By avoiding frontal collisions that blind the detector, it is possible to observe the photons that interact and "bounce" off each other.

Studying more than 4 billion events recorded in 2015, the ATLAS collaboration has found 13 candidates for light-by-light scattering in which the lead ions didn't collide, but came close enough for their electromagnetic fields to interact. Their signature—two photons—is the diametric opposite of the tremendous complexity expected from lead nuclei collisions.

Achieving this feat required the development a new, highly sensitive "trigger" for the ATLAS detector.

This study has other implications for theories. The photon scattering process predicts that photons exchange virtual charged particles, either known or unkown. The physicists from ATLAS will continue their research in 2018 to refine the accuracy of their results, and possibly detect the hints of a new physics.