Rise time of proton cut-off energy in 2D and 3D PIC simulations

TL;DR

This paper introduces empirical laws describing the rise time of proton cut-off energy in 2D and 3D PIC simulations of laser-driven proton acceleration, enabling more accurate and computationally efficient analysis of simulation results.

Contribution

It proposes two empirical laws for the rise time of proton cut-off energy in 2D and 3D PIC simulations, validated by simulation data and useful for reducing computational costs.

Findings

The proposed laws fit simulation data well.

2D simulations can approximate 3D results.

Cut-off energy depends on target thickness and incidence angle.

Abstract

The Target Normal Sheath Acceleration (TNSA) regime for proton acceleration by laser pulses is experimentally consolidated and fairly well understood. However, uncertainties remain in the analysis of particle-in-cell (PIC) simulation results. The energy spectrum is exponential with a cut-off, but the maximum energy depends on the simulation time, following different laws in two and three dimensional (2D, 3D) PIC simulations, so that the determination of an asymptotic value has some arbitrariness. We propose two empirical laws for rise time of the cut-off energy in 2D and 3D PIC simulations, suggested by a model in which the proton acceleration is due to a surface charge distribution on the target rear side. The kinetic energy of the protons that we obtain follows two distinct laws, which appear to be nicely satisfied by PIC simulations. The laws depend on two parameters: the scaling time, at which the energy starts to rise, and the asymptotic cut-off energy. The values of the cut-off energy, obtained by fitting the 2D and 3D simulations for the same target and laser pulse, are comparable. This suggests that parametric scans can be performed with 2D simulations, since 3D ones are computationally very expensive. In this paper, the simulations are carried out for $a_0=3$ with the PIC code ALaDyn by changing the target thickness $L$ and the incidence angle $\alpha$. A monotonic dependence, on $L$ for normal incidence and on $\alpha$ for fixed $L$, is found, as in the experimental results for high temporal contrast pulses.