Birth process of electron-positron pairs inside supercritical fields

Abstract

We propose a computational quantum field theoretical approach to obtain a microscopic insight into the creation process of electrons and positrons as well as their subsequent motion inside a supercritical external field with space-time resolution. A machine-learning–based method permits us to address fundamental questions such as where inside the interaction region the particles are being created and what their initial velocity distribution is. It suggests that the particles' most likely birth positions change in time during the dynamics. At early times the particles' birth density is roughly proportional to the square of the force field, but in the long-time and steady state production regime their possible birth locations narrow down significantly. Counterintuitively, this leads for longer times to the occurrence of “birth-free” zones within the field, where particles are no longer created even though the electric field is maximal there. The genetic-programming–based symbolic regression algorithms first learn multiple sequences of partially dressed positronic spatial probability densities as training data and then exploit their features as a function of the dressing strength in order to predict the particles' true distribution in space and momentum.