pith. machine review for the scientific record. sign in

arxiv: 2605.12132 · v1 · submitted 2026-05-12 · ⚛️ physics.plasm-ph

Recognition: no theorem link

High-order exponential solver method for particle-in-cell simulations in cylindrical geometry

Nasr A.M. Hafz, Szil\'ard Majorosi, Zsolt L\'ecz

Pith reviewed 2026-05-13 03:17 UTC · model grok-4.3

classification ⚛️ physics.plasm-ph
keywords particle-in-cell simulationscylindrical geometryfinite difference methodsexponential time integratorlaser wakefield accelerationMaxwell equationsnumerical plasma physics
0
0 comments X

The pith

High-order finite differences with exponential integration achieve spectral-level accuracy in cylindrical PIC simulations without basis transformations.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper develops a real-space alternative to spectral cylindrical PIC codes by combining high-order staggered finite differences with an exponential time integrator for Maxwell equations and particle motion. It also introduces special near-axis particle handling and an exponential propagator for the laser envelope potential. This setup is tested on laser propagation in vacuum and plasma plus nonlinear wakefield electron injection, showing close agreement with full 3D runs and with the FBPIC spectral code, though the spectral code yields slightly less energetic electrons and smoother on-axis distributions. The approach matters because it keeps the computational savings of cylindrical geometry while avoiding the implementation overhead of special basis functions.

Core claim

We present a finite-difference exponential time-domain method that solves the cylindrically symmetric Maxwell equations and particle dynamics at high accuracy using locally staggered high-order differences, near-axis particle corrections, and an exponential integrator for the laser envelope; benchmarks confirm convergence and good agreement with 3D and spectral results without any transformation to special basis functions.

What carries the argument

High-order staggered finite differences combined with an exponential time integrator applied to the cylindrical Maxwell equations and particle motion, together with an exponential envelope propagator.

If this is right

  • Cylindrical simulations can reach accuracy close to full 3D for laser wakefield acceleration at reduced cost.
  • The envelope exponential propagator maintains high fidelity for laser propagation in underdense plasma.
  • Near-axis particle handling produces distributions comparable to spectral methods in most respects.
  • The method converges reliably across vacuum and plasma benchmarks, supporting use in nonlinear regimes.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Existing finite-difference PIC codes could adopt the scheme with relatively modest changes compared with adding spectral transforms.
  • Observed differences in electron energies versus FBPIC point to possible trade-offs in axis treatment that warrant targeted diagnostics.
  • The real-space formulation may allow easier addition of extra physics modules without redesigning basis expansions.

Load-bearing premise

High-order staggered finite differences plus the exponential integrator and near-axis particle handling can reproduce the physics of cylindrical Maxwell equations and particle motion without the errors that spectral methods avoid by construction.

What would settle it

If the error in electric-field or particle-energy spectra for vacuum laser propagation or nonlinear wakefield injection grows faster than the expected convergence order with grid refinement, or if near-axis electron distributions deviate markedly from 3D reference runs.

Figures

Figures reproduced from arXiv: 2605.12132 by Nasr A.M. Hafz, Szil\'ard Majorosi, Zsolt L\'ecz.

Figure 1
Figure 1. Figure 1: Results of density depositions of a radially uniform density [PITH_FULL_IMAGE:figures/full_fig_p007_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Error characteristics of a laser pulse of [PITH_FULL_IMAGE:figures/full_fig_p011_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Relative difference between c and the group velocity vg of the pulse centroid in a plasma of 10−3nc density for different N = λ/∆z resolu￾tions in cylindrical geometry. The black line represents the analytical predic￾tion (5.634 × 10−4 ), the blue, green lines (circle markers) correspond to sim￾ulations using azimuthal modes decomposition with 5 basis functions, the rest propagates the laser potentials (A,… view at source ↗
Figure 4
Figure 4. Figure 4: Plot of the linear wakefields Ez (top panels), Ey (bottom panels) generated by a laser pulse of λ = 0.8µm after 100µm propagation in an underdense plasma in cylindrical geometry. We show the colormap of these fields in (a), (c) in the x = 0 plane where we also have indicated the position of the laser envelope and a line cross section. We show the latter on panels (b), (d) to compare the accuracy of the wak… view at source ↗
Figure 5
Figure 5. Figure 5: Comparison of simulations with different configurations (a)-(e). In the first column the electron density distributions are shown with logarithmic color scale (log10[ne(cm−3 )]), in the second and third columns the pz-px and pz-py distributions are presented, in the plane of laser’s polarization plane and in the orthogonal plane, respectively. We also show the Ex − cBy field on (d4) where the numerical Che… view at source ↗
read the original abstract

Recent developments in high peak-power table-top laser systems reaching highly relativistic light intensities have led to significant advances in laser-driven particle acceleration schemes (mainly the laser wakefield acceleration, LWFA) that heavily rely on particle-in-cell (PIC) simulations for the microscopic understanding of the acceleration process. Efficient algorithms have been developed by taking advantage of the cylindrical geometry of the laser-plasma acceleration interaction, which reduces the computational and memory costs of these simulations, but with the trade-off of reduced accuracy compared to the 3D simulations. The most successful solution solves the Maxwell equations on a Fourier-Bessel spectral basis in this geometry, as used by the well-known FBPIC code. In this work, we present a solution that is a real-space equivalent of the latter using the finite difference exponential time-domain method. Spatially, we represent the derivatives with high-order staggered finite differences locally and address issues of the near-axis particle representation. Additionally, we also develop an exponential solution to propagate the laser envelope potential with high accuracy in the cylindrically symmetric PIC model. We show that this method provides a very high accuracy without relying on a transformation to special basis functions. We verified the accuracy and the convergence of these methods in various benchmarks involving laser propagation in vacuum and in underdense plasma. Electron injection in the non-linear laser wakefield regime has also been simulated and the results are compared with 3D simulations, and to the cylindrical spectral solution of FBPIC. We found good agreement between these methods; however, the spectral solution resulted in less energetic electrons and a smoother spatial distribution near the cylindrical axis.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript presents a real-space high-order staggered finite-difference exponential time integrator for Maxwell's equations and particle motion in cylindrical PIC simulations. It includes specialized near-axis particle handling and an exponential solver for the laser envelope potential, claiming very high accuracy without transformation to Fourier-Bessel basis functions. Accuracy is verified through benchmarks on vacuum laser propagation, underdense plasma, and nonlinear LWFA electron injection, with comparisons to full 3D simulations and the spectral FBPIC code showing overall good agreement, though with noted differences in electron energies and near-axis smoothness.

Significance. If the accuracy and convergence claims hold under quantitative scrutiny, the method supplies a practical real-space alternative to spectral cylindrical PIC codes. This could reduce implementation barriers for users preferring finite-difference frameworks while maintaining efficiency gains from cylindrical symmetry in laser-plasma acceleration studies.

major comments (2)
  1. [Abstract and benchmarks] Abstract and benchmarks section: The central claim of 'very high accuracy' without special basis functions is load-bearing, yet the reported differences (spectral solution yields less energetic electrons and smoother near-axis distribution) are presented only qualitatively. No L2 error norms, radial convergence rates, or order-of-accuracy tests against an exact cylindrical solution are provided to demonstrate that residual grid artifacts are suppressed below spectral levels.
  2. [Near-axis particle representation] Near-axis particle representation section: The finite-difference handling of particles and fields near r=0 is described as addressing singularities, but the manuscript provides no explicit stencil derivation, conservation proof, or truncation-error analysis for this fix. This is critical because any local error here directly impacts the claim that the real-space scheme reproduces the physics without the artifacts spectral bases avoid.
minor comments (2)
  1. [Figures and benchmarks] Figure captions and benchmark descriptions should include grid resolution, time step, and exact laser/plasma parameters to enable direct reproduction of the reported agreement levels.
  2. [Methods] Notation for the staggered finite-difference operators and exponential propagator should be defined once in a dedicated methods subsection rather than inline.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed comments. We address each major point below and will revise the manuscript accordingly to provide the requested quantitative analyses and derivations.

read point-by-point responses

  1. Referee: [Abstract and benchmarks] Abstract and benchmarks section: The central claim of 'very high accuracy' without special basis functions is load-bearing, yet the reported differences (spectral solution yields less energetic electrons and smoother near-axis distribution) are presented only qualitatively. No L2 error norms, radial convergence rates, or order-of-accuracy tests against an exact cylindrical solution are provided to demonstrate that residual grid artifacts are suppressed below spectral levels.

    Authors: We agree that quantitative error measures would strengthen the central claim. The manuscript already states that accuracy and convergence were verified in the benchmarks, but we will add explicit L2 error norms and radial convergence rates for the vacuum laser propagation case against the known analytical solution. For the underdense plasma and nonlinear LWFA injection benchmarks, where no exact cylindrical solution exists, we will report quantitative differences (e.g., relative errors in electron energy spectra and beam emittance) between our method, FBPIC, and 3D simulations. These additions will be included in a revised benchmarks section. revision: yes

  2. Referee: [Near-axis particle representation] Near-axis particle representation section: The finite-difference handling of particles and fields near r=0 is described as addressing singularities, but the manuscript provides no explicit stencil derivation, conservation proof, or truncation-error analysis for this fix. This is critical because any local error here directly impacts the claim that the real-space scheme reproduces the physics without the artifacts spectral bases avoid.

    Authors: We acknowledge that the current description of the near-axis treatment is insufficiently detailed. In the revised manuscript we will add an appendix or subsection containing the explicit derivation of the specialized staggered finite-difference stencils for both fields and particles at r=0, a truncation-error analysis confirming that the designed order of accuracy is retained, and a demonstration (analytic or numerical) of local conservation properties such as charge and momentum. These additions will directly support the claim that the real-space approach avoids near-axis artifacts without requiring a spectral basis. revision: yes

Circularity Check

0 steps flagged

No circularity: method uses standard FD/exponential integrators with external benchmark verification

full rationale

The paper describes a real-space finite-difference exponential time integrator for cylindrical PIC, using high-order staggered differences and near-axis fixes. It verifies accuracy via direct comparisons to 3D simulations and the independent FBPIC spectral code on vacuum propagation, underdense plasma, and LWFA injection benchmarks. No equations reduce to self-definition, no fitted parameters are relabeled as predictions, and no load-bearing uniqueness or ansatz is imported solely via self-citation. The claim of high accuracy without special bases rests on these external comparisons rather than internal construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The method rests on standard numerical techniques for Maxwell equations in cylindrical coordinates and particle pusher algorithms; no new physical entities or heavily fitted parameters are introduced.

axioms (2)
  • standard math Maxwell's equations hold in cylindrical coordinates with azimuthal symmetry for the laser-plasma interaction
    Invoked as the basis for the PIC model throughout the abstract.
  • domain assumption High-order staggered finite differences provide accurate spatial derivatives near the axis when combined with appropriate particle representation
    Central to the real-space approach; the abstract notes addressing near-axis issues but does not detail the exact scheme.

pith-pipeline@v0.9.0 · 5601 in / 1303 out tokens · 69086 ms · 2026-05-13T03:17:13.364478+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

32 extracted references · 32 canonical work pages

  1. [1]

    Strickland, G

    D. Strickland, G. Mourou, Compression of am- plified chirped optical pulses, Optics Commu- nications 55 (6) (1985) 447–449.doi:https: //doi.org/10.1016/0030-4018(85)90151-8. URLhttps://www.sciencedirect.com/science/ article/pii/0030401885901518 15

    work page doi:10.1016/0030-4018(85)90151-8 1985

  2. [2]

    Tajima, J

    T. Tajima, J. M. Dawson, Laser electron ac- celerator, Phys. Rev. Lett. 43 (1979) 267–270. doi:10.1103/PhysRevLett.43.267. URLhttps://link.aps.org/doi/10.1103/ PhysRevLett.43.267

    work page doi:10.1103/physrevlett.43.267 1979

  3. [3]

    Esarey, C

    E. Esarey, C. B. Schroeder, W. P. Leemans, Physics of laser-driven plasma-based electron ac- celerators, Rev. Mod. Phys. 81 (2009) 1229–1285. doi:10.1103/RevModPhys.81.1229. URLhttps://link.aps.org/doi/10.1103/ RevModPhys.81.1229

    work page doi:10.1103/revmodphys.81.1229 2009

  4. [4]

    W. Lu, M. Tzoufras, C. Joshi, F. S. Tsung, W. B. Mori, J. Vieira, R. A. Fonseca, L. O. Silva, Gener- ating multi-GeV electron bunches using single stage laser wakefield acceleration in a 3d nonlinear regime, Phys. Rev. ST Accel. Beams 10 (2007) 061301. doi:10.1103/PhysRevSTAB.10.061301. URLhttps://link.aps.org/doi/10.1103/ PhysRevSTAB.10.061301

    work page doi:10.1103/physrevstab.10.061301 2007

  5. [5]

    Albert, A

    F. Albert, A. G. R. Thomas, Applications of laser wakefield accelerator-based light sources, Plasma Physics and Controlled Fusion 58 (10) (2016) 103001. doi:10.1088/0741-3335/58/10/103001. URLhttps://doi.org/10.1088/0741-3335/58/ 10/103001

    work page doi:10.1088/0741-3335/58/10/103001 2016

  6. [6]

    C. K. Birdsall, A. B. Langdon, Plasma Physics via Com- puter Simulation, CRC press, 1991.doi:https://doi. org/10.1201/9781315275048

    work page doi:10.1201/9781315275048 1991

  7. [7]

    R. A. Fonseca, L. O. Silva, F. S. Tsung, V . K. Decyk, W. Lu, C. Ren, W. B. Mori, S. Deng, S. Lee, T. Kat- souleas, J. C. Adam, Osiris: A three-dimensional, fully relativistic particle in cell code for modeling plasma based accelerators, in: P. M. A. Sloot, A. G. Hoekstra, C. J. K. Tan, J. J. Dongarra (Eds.), Computational Science — ICCS 2002, Springer Be...

    work page 2002

  8. [8]

    P. L. Pritchett, Particle-in-cell simulation of plasmas—a tutorial, Space Plasma Simulation (2003) 1–24

    work page 2003

  9. [9]

    T. D. Arber, K. Bennett, C. S. Brady, A. Lawrence- Douglas, M. G. Ramsay, N. J. Sircombe, P. Gillies, R. G. Evans, H. Schmitz, A. R. Bell, C. P. Ridgers, Contempo- rary particle-in-cell approach to laser-plasma modelling, Plasma Physics and Controlled Fusion 57 (11) (2015) 113001.doi:10.1088/0741-3335/57/11/113001. URLhttps://doi.org/10.1088/0741-3335/57/...

    work page doi:10.1088/0741-3335/57/11/113001 2015

  10. [10]

    M. A. Belyaev, PICsar: A 2.5d axisymmet- ric, relativistic, electromagnetic, particle in cell code with a radiation absorbing boundary, New Astronomy 36 (2015) 37–49.doi:https: //doi.org/10.1016/j.newast.2014.09.006. URLhttps://www.sciencedirect.com/science/ article/pii/S1384107614001407

    work page doi:10.1016/j.newast.2014.09.006 2015

  11. [11]

    Derouillat, A

    J. Derouillat, A. Beck, F. Pérez, T. Vinci, M. Chiaramello, A. Grassi, M. Flé, G. Bouchard, I. Plotnikov, N. Au- nai, J. Dargent, C. Riconda, M. Grech, Smilei : A collaborative, open-source, multi-purpose particle-in- cell code for plasma simulation, Computer Physics Communications 222 (2018) 351–373.doi:https: //doi.org/10.1016/j.cpc.2017.09.024. URLhttp...

    work page doi:10.1016/j.cpc.2017.09.024 2018

  12. [12]

    Lifschitz, X

    A. Lifschitz, X. Davoine, E. Lefebvre, J. Faure, C. Rechatin, V . Malka, Particle-in-cell modelling of laser–plasma interaction using Fourier decomposition, Journal of Computational Physics 228 (5) (2009) 1803– 1814.doi:https://doi.org/10.1016/j.jcp.2008. 11.017. URLhttps://www.sciencedirect.com/science/ article/pii/S0021999108005950

    work page doi:10.1016/j.jcp.2008 2009

  13. [13]

    Davidson, A

    A. Davidson, A. Tableman, W. An, F. Tsung, W. Lu, J. Vieira, R. Fonseca, L. Silva, W. Mori, Implementation of a hybrid particle code with a PIC description in r–z and a gridless description in phi into osiris, Journal of Computational Physics 281 (2015) 1063–1077.doi: https://doi.org/10.1016/j.jcp.2014.10.064. URLhttps://www.sciencedirect.com/science/ art...

    work page doi:10.1016/j.jcp.2014.10.064 2015

  14. [14]

    R. Lehe, M. Kirchen, I. A. Andriyash, B. B. Godfrey, J.-L. Vay, A spectral, quasi-cylindrical and dispersion- free particle-in-cell algorithm, Computer Physics Communications 203 (2016) 66–82.doi:https: //doi.org/10.1016/j.cpc.2016.02.007. URLhttps://www.sciencedirect.com/science/ article/pii/S0010465516300224

    work page doi:10.1016/j.cpc.2016.02.007 2016

  15. [15]

    Majorosi, N

    S. Majorosi, N. A. Hafz, Z. Lécz, High-order exponential solver method for particle-in-cell simulations, Computer Physics Communications 322 (2026) 110054.doi: https://doi.org/10.1016/j.cpc.2026.110054. URLhttps://www.sciencedirect.com/science/ article/pii/S0010465526000366

    work page doi:10.1016/j.cpc.2026.110054 2026

  16. [16]

    Leforestier, R

    C. Leforestier, R. Bisseling, C. Cerjan, M. Feit, R. Fries- ner, A. Guldberg, A. Hammerich, G. Jolicard, W. Kar- rlein, H.-D. Meyer, N. Lipkin, O. Roncero, R. Kosloff, A comparison of different propagation schemes for the time dependent Schrödinger equation, Journal of Com- putational Physics 94 (1) (1991) 59–80.doi:https: //doi.org/10.1016/0021-9991(91)9...

    work page doi:10.1016/0021-9991(91)90137-a 1991

  17. [17]

    Castro, M

    A. Castro, M. A. L. Marques, A. Rubio, Propagators for the time-dependent Kohn–Sham equations, The Journal of Chemical Physics 121 (8) (2004) 3425–3433.arXiv: 16 https://pubs.aip.org/aip/jcp/article-pdf/ 121/8/3425/19313892/3425\_1\_online.pdf, doi:10.1063/1.1774980. URLhttps://doi.org/10.1063/1.1774980

    work page doi:10.1063/1.1774980 2004

  18. [18]

    A. D. Bandrauk, H. LU, Exponential propagators (integra- tors) for the time-dependent Schrödinger equation, Jour- nal of Theoretical and Computational Chemistry 12 (06) (2013) 1340001.doi:https://doi.org/10.1142/ S0219633613400014

    work page 2013

  19. [19]

    B. M. Cowan, D. L. Bruhwiler, E. Cormier-Michel, E. Esarey, C. G. Geddes, P. Messmer, K. M. Paul, Characteristics of an envelope model for laser–plasma accelerator simulation, Journal of Computational Physics 230 (1) (2011) 61–86.doi:https: //doi.org/10.1016/j.jcp.2010.09.009. URLhttps://www.sciencedirect.com/science/ article/pii/S0021999110005036

    work page doi:10.1016/j.jcp.2010.09.009 2011

  20. [20]

    Benedetti, C

    C. Benedetti, C. B. Schroeder, C. G. R. Geddes, E. Esarey, W. P. Leemans, An accurate and efficient laser-envelope solver for the modeling of laser-plasma accelerators, Plasma Physics and Controlled Fusion 60 (1) (2017) 014002.doi:10.1088/1361-6587/aa8977. URLhttps://dx.doi.org/10.1088/1361-6587/ aa8977

    work page doi:10.1088/1361-6587/aa8977 2017

  21. [21]

    Terzani, P

    D. Terzani, P. Londrillo, A fast and accurate nu- merical implementation of the envelope model for laser–plasma dynamics, Computer Physics Com- munications 242 (2019) 49–59.doi:https: //doi.org/10.1016/j.cpc.2019.04.007. URLhttps://www.sciencedirect.com/science/ article/pii/S0010465519301195

    work page doi:10.1016/j.cpc.2019.04.007 2019

  22. [22]

    W. H. Press, Numerical recipes 3rd edition: The art of scientific computing, Cambridge university press, 2007

    work page 2007

  23. [23]

    Constantinescu, S

    G. Constantinescu, S. Lele, A highly accurate technique for the treatment of flow equations at the polar axis in cylindrical coordinates using series expansions, Journal of Computational Physics 183 (1) (2002) 165–186. doi:https://doi.org/10.1006/jcph.2002.7187. URLhttps://www.sciencedirect.com/science/ article/pii/S0021999102971871

    work page doi:10.1006/jcph.2002.7187 2002

  24. [24]

    B. B. Godfrey, J.-L. Vay, Numerical stability of relativistic beam multidimensional PIC simulations employing the esirkepov algorithm, Journal of Com- putational Physics 248 (2013) 33–46.doi:https: //doi.org/10.1016/j.jcp.2013.04.006. URLhttps://www.sciencedirect.com/science/ article/pii/S0021999113002556

    work page doi:10.1016/j.jcp.2013.04.006 2013

  25. [25]

    J.-X. Li, Y . I. Salamin, K. Z. Hatsagortsyan, C. H. Keitel, Fields of an ultrashort tightly focused laser pulse, J. Opt. Soc. Am. B 33 (3) (2016) 405–411. doi:10.1364/JOSAB.33.000405. URLhttp://opg.optica.org/josab/abstract. cfm?URI=josab-33-3-405

    work page doi:10.1364/josab.33.000405 2016

  26. [26]

    Majorosi, Z

    S. Majorosi, Z. Lécz, D. Papp, C. Kamperidis, N. A. M. Hafz, Numerical representation of tightly focused ultra- short laser pulses, J. Opt. Soc. Am. B 40 (3) (2023) 551–559.doi:10.1364/JOSAB.481864. URLhttps://opg.optica.org/josab/abstract. cfm?URI=josab-40-3-551

    work page doi:10.1364/josab.481864 2023

  27. [27]

    van Dijk, F

    W. van Dijk, F. M. Toyama, Numerical solutions of the Schrödinger equation with source terms or time- dependent potentials, Phys. Rev. E 90 (2014) 063309. doi:10.1103/PhysRevE.90.063309. URLhttps://link.aps.org/doi/10.1103/ PhysRevE.90.063309

    work page doi:10.1103/physreve.90.063309 2014

  28. [28]

    Blanes, F

    S. Blanes, F. Casas, J. Oteo, J. Ros, The Magnus expansion and some of its applications, Physics Reports 470 (5) (2009) 151–238.doi:https: //doi.org/10.1016/j.physrep.2008.11.001. URLhttps://www.sciencedirect.com/science/ article/pii/S0370157308004092

    work page doi:10.1016/j.physrep.2008.11.001 2009

  29. [29]

    Esarey, P

    E. Esarey, P. Sprangle, M. Pilloff, J. Krall, Theory and group velocity of ultrashort, tightly focused laser pulses, J. Opt. Soc. Am. B 12 (9) (1995) 1695–1703. doi:10.1364/JOSAB.12.001695. URLhttp://opg.optica.org/josab/abstract. cfm?URI=josab-12-9-1695

    work page doi:10.1364/josab.12.001695 1995

  30. [30]

    S. G. Johnson, Notes on perfectly matched layers (PMLs), arXiv preprint arXiv:2108.05348 (2021)

    work page arXiv 2021

  31. [31]

    Barucq, M

    H. Barucq, M. Fontes, Well-posedness and exponential stability of Maxwell-like systems coupled with strongly absorbing layers, Journal de Mathématiques Pures et Appliquées 87 (3) (2007) 253–273.doi:https: //doi.org/10.1016/j.matpur.2007.01.001. URLhttps://www.sciencedirect.com/science/ article/pii/S0021782407000165

    work page doi:10.1016/j.matpur.2007.01.001 2007

  32. [32]

    A. V . Higuera, J. R. Cary, Structure-preserving second- order integration of relativistic charged particle tra- jectories in electromagnetic fields, Physics of Plasmas 24 (5) (2017) 052104.arXiv:https://pubs. aip.org/aip/pop/article-pdf/doi/10.1063/ 1.4979989/15988441/052104\_1\_online.pdf, doi:10.1063/1.4979989. URLhttps://doi.org/10.1063/1.4979989 17 F...

    work page doi:10.1063/1.4979989 2017