An Overview of Relativistic Particle Pushers and their Extension to Arbitrary Order Accuracy

H. Schmitz · 2026

Read An Overview of Relativistic Particle Pushers and their Extension to Arbitrary Order Accuracy by H. Schmitz (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-06509.txt.

Introduction

Schmitz tests particle pushers for relativistic trajectories in PIC plasma simulations, where correct integration and error care matter for judging run truth. Many studied systems contain fast particles. Various new schemes for their paths have appeared over recent decades to improve accuracy in specific cases. The article presents a full comparison focused on explicit plans, finding a key class wide to arbitrary high order, plus fourth-order variants against their second-order counterparts.

Catalog record

CREATORH. Schmitz
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE69899
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.06509

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

An Overview of Relativistic Particle Pushers and their Extension to Arbitrary Order Accuracy H. Schmitz Abstract Particle in Cell (PIC) simulations have become a vital tool for the investigation of kinetic processes in plasma physics. Many of the systems investigated with PIC simulations contain particles with relativistic velocities. The correct integration and the knowledge of possible sources of errors in relativistic particle trajectories is of importance to accurately judge the validity of the simulation results. Over the past few decades, various new integration schemes for relativistic particle trajectories in PIC simulations have been proposed. These are aimed at improving numerical accuracy in specific scenarios. This article presents a comprehensive comparison of particle pushers with a focus on explicit schemes. An important class of these schemes is found to be generalisable to arbitrary high order. A comparison of the fourth order variants of these schemes with their second order counterpart is also presented. 1 Introduction Since its inception, the Particle In Cell (PIC) [ 9 , 5 ] method has become an indispensable tool in the investigation of kinetic phenomena in plasma physics. The PIC method consists of two main components, a field solver that calculates the electromagnetic field on a grid, and an integration scheme for the particle orbits, often called the "particle pusher", that calculates the trajectories of the simulation particles. These two components are joined together by suitable interpolation schemes that, on one hand, interpolate the forces of the electromagnetic field from the grid onto the particle positions and, on the other hand, interpolate the particle charges or currents back onto the grid. From the beginning, the PIC technique has been applied to relativistic scenarios [ 5 ] . Nowadays relativistic PIC simulations are fundamental in fields ranging from astrophysics [ 28 , 24 ] and space and planetary physics [ 6 , 12 , 18 ] to fusion research and high energy density research [ 33 , 20 , 32 ] . Today, many large scale relativistic simulation codes exist, such as Osiris [ 13 ] , EPOCH [ 1 ] , SMILEI [ 11 ] , WarpX [ 38 ] , and others. Relativistic PIC simulations necessitate the need for relativistic particle pushers. The integration method proposed by Boris [ 4 ] has generally become the dominant scheme for most PIC simulation codes. The Boris scheme is easy to implement, has second order accuracy and very good long time stability, often attributed to the volume preserving nature of the scheme [ 29 ] . More recently however, it was found that the Boris scheme has some shortcomings in relativistically strong electromagnetic fields. Arefiev et al [ 2 ] found that constants of motion were only poorly conserved for particles accelerated in a high-intensity plane electromagnetic wave. A sub-cycling scheme was proposed based on the instantaneous strength of the magnetic field. Vay [ 39 ] showed that the calculated trajectory of a relativistic particle propagating in a constant crossed electromagnetic field, such that the total force on the particle vanishes, can deviate from the ideal straight trajectory. He proposed an alternative scheme specifically tailored to produce the correct trajectory in this scenario. As a response to the numerical inaccuracies of the Boris scheme in the relativistic regime, numerous alternative integration schemes have been proposed during the last decade [ 19 , 39 , 26 , 46 , 7 , 23 , 16 , 10 , 48 , 31 ] . Reviews of techniques in PIC simulations have a history almost as old as the PIC simulation scheme itself [ 8 , 3 ] . Today, many reviews of PIC methods exist focusing on the application in specific fields [ 34 , 28 , 24 , 30 , 41 ] . There exist also comprehensive reviews and text books on PIC simulations covering topics such as the field solver, the particle interpolation and deposition schemes and the particle pusher [ 36 , 40 , 14 ] . Some recent comparisons focusing specifically on different particle integration schemes can be found in Ripperda et al [ 31 ] and Yasir & Saxena [ 42 ] . Ripperda et al [ 31 ] compares the Boris pusher with the integrators proposed by Vay [ 39 ] and Higuera & Cary [ 19 ] together with their new implicit midpoint method. However, the comparison provides few quantitative results and is limited to a small selection of currently available schemes. Yasir & Saxena [ 42 ] also compare the integrators by Boris, Vay, and Higuera & Cary but focus mostly on the computational efficiency of their implementation. This article aims to provide a comprehensive overview of the numerical performance of integration schemes in the relativistic regime. The aim is to classify and compare the various schemes that have been proposed over the last two decades. All schemes considered here are tested in a variety of scenarios and, wherever possible, quantitative measures of the numerical accuracy are given and compared. The hope is to provide guidance to developers and users of simulation codes alike as to which schemes are better suited for a specific scenario. The main focus of the comparison is explicit schemes. These are most often preferred in implementations of the PIC method due to their better computational performance. However, one implicit scheme by Ripperda et al [ 31 ] will be included to provide a benchmark. The paper is organised as follows. Section 2 introduces the integration schemes used throughout. A brief numerical outline is presented for some of the schemes. For the remaining schemes, the reader is referred to the respective publications. Section 3 presents the test cases and describes the results. In section 4 a class of "Boris-like" schemes is identified that can be extended to higher order using a method proposed by Yoshida [ 45 ] . Fourth order schemes are implemented and tested against their second order counterparts for a selection of tests. Finally, section 5 discusses the results. 2 Outline of Integration Schemes The relativistic equations of motion for a charged particle in an electromagnetic field are given by \displaystyle\frac{d\mathbf{x}}{dt} \displaystyle=\frac{\mathbf{u}}{\gamma} (1) \displaystyle\frac{d\mathbf{u}}{dt} \displaystyle=\frac{q}{m}\left(\mathbf{E}+\frac{1}{\gamma}\mathbf{u}\times\mathbf{B}\right) (2) Here \mathbf{x} is the particle’s position and \mathbf{u}=\gamma\mathbf{v} are the spatial components of the four-velocity, with the velocity, \mathbf{v} , and the Lorentz factor, \gamma . The Lorentz factor can be expressed as a function of the magnitude of the relativistic spatial velocity, \gamma(u)=(1+u^{2}/c^{2})^{1/2} . In the following, \mathbf{u} will simply be denoted as the relativistic velocity. The schemes considered in this work can roughly be split into two main types. Type I will denote time centered explicit second order schemes. Type II will denote schemes integrating the trajectory in the particle’s proper time. Finally, one implicit scheme by Ripperda et al [ 31 ] will be also considered for comparison. 2.1 Type I: time centered explicit second order schemes An important subset of time-centered second order schemes, which includes the classic Boris scheme [ 4 ] , the Higuera & Cary scheme [ 19 ] , and Vay’s scheme [ 38 ] , can be expressed as [ 19 ] , \displaystyle\mathbf{x}^{(n+1/2)}-\mathbf{x}^{(n)} \displaystyle=\Delta t\frac{\mathbf{u}^{(n)}}{2\gamma(u^{(n)})} (3) \displaystyle\mathbf{u}^{(n+1)}-\mathbf{u}^{(n)} \displaystyle=\Delta t\frac{q}{m}\left(\mathbf{E}^{(n+1/2)}+\bar{\mathbf{v}}\times\mathbf{B}^{(n+1/2)}\right) (4) \displaystyle\mathbf{x}^{(n+1)}-\mathbf{x}^{(n+1/2)} \displaystyle=\Delta t\frac{\mathbf{u}^{(n+1)}}{2\gamma(u^{(n+1)})}. (5) Here, the superscript denotes the time

Read the complete text