Estimation of direct laser acceleration in laser wakefield accelerators using particle-in-cell simulations

Abstract: Many current laser wakefield acceleration (LWFA) experiments are carried out in a regime where the laser pulse length is on the order of or longer than the wake wavelength and where ionization injection is employed to inject electrons into the wake. In these experiments, the trapped electrons will co-propagate with the longitudinal wakefield and the transverse laser field. In this scenario, the electrons can gain a significant amount of energy from both the direct laser acceleration (DLA) mechanism as well as the usual LWFA mechanism. Particle-in-cell (PIC) codes are frequently used to discern the relative contribution of these two mechanisms. However, if the longitudinal resolution used in the PIC simulations is inadequate, it can produce numerical heating that can overestimate the transverse motion, which is important in determining the energy gain due to DLA. We have therefore carried out a systematic study of this LWFA regime by varying the longitudinal resolution of PIC simulations from the standard, best-practice resolution of 30 points per laser wavelength to four times that value and then examining the energy gain characteristics of both the highest-energy electrons and the bulk electrons. By calculating the contribution of DLA to the final energies of the electrons produced from the LWFA, we find that although the transverse momentum and oscillation radii are over-estimated in the lower-resolution simulations, this over-estimation does not lead to artificial energy gain by DLA. Rather, the DLA contribution to the highest-energy electrons is larger in the higher-resolution cases because the DLA resonance is better maintained. Thus, even at the highest longitudinal resolutions, DLA contributes a significant portion of the energy gained by the highest-energy electrons and also contributes to accelerating the bulk of the charge in the electron beam produced by the LWFA.