This paper formulates a new particle-in-cell method for the Vlasov-Maxwell system. Under the Lorenz gauge condition, Maxwell's equations for the electromagnetic fields can be written as a collection of scalar and vector wave equations. The use of potentials for the fields motivates the adoption of a Hamiltonian formulation for particles that employs the generalized momentum. The resulting updates for particles require only knowledge of the fields and their spatial derivatives. An analytical method for constructing these spatial derivatives is presented that exploits the underlying integral solution used in the field solver for the wave equations. Moreover, these derivatives are shown to converge at the same rate as the fields in the both time and space. The field solver we consider in this work is first-order accurate in time and fifth-order accurate in space and belongs to a larger class of methods which are unconditionally stable, can address geometry, and leverage fast summation methods for efficiency. We demonstrate the method on several well-established benchmark problems, and the efficacy of the proposed formulation is demonstrated through a comparison with standard methods presented in the literature. The new method shows mesh-independent numerical heating properties even in cases where the plasma Debye length is close to the grid spacing. The use of high-order spatial approximations in the new method means that fewer grid points are required in order to achieve a fixed accuracy. Our results also suggest that the new method can be used with fewer simulation particles per cell compared to standard explicit methods, which permits further computational savings.
Introduction
This paper introduces a new particle-in-cell (PIC) method for the Vlasov-Maxwell (VM) system that is based on a potential formulation in which the particle updates are advanced through a Hamiltonian formulation. The new method shows several advantages over traditional PIC methods, namely: mesh-independent numerical heating; stability and refinement of solutions even when the mesh spacing is comparable to the Debye length (i.e., λD/Δx≈1); enhanced ability to capture symmetries that pose difficulties for standard PIC methods; and can be used in problems with complex geometry without resorting to cut-cells and staircase approximations. The method is stable even when the mesh spacing exceeds the Debye length (i.e., λD/Δx>1), although accuracy may become an issue. The ability to use λD/Δx>1 is important when simulating dense plasmas and implies that the method requires fewer simulation particles than comparable explicit PIC methods based on finite-differences for a given accuracy. Indeed, for benchmark test problems, which simulate plasmas in both periodic and bounded domains, the new method outperforms conventional explicit PIC methods that use staggered meshes. The field solver used in the new method is non-staggered and is fifth-order accurate in space and globally first-order accurate in time. Because we adopt the generalized momentum formulation for the particle push, we do not require temporal derivatives of the fields. For the problems considered in this work, the first-order time accuracy of the fields has been adequate, especially given that the spatial derivatives are computed to high-order accuracy using “analytical” expressions. As our goal is to eventually move to complex domains, the new method makes use of bilinear (area) weightings for mapping the charge and current density to the mesh. It is well-known that this approach leads to charge conservation issues [1]. However, in benchmark problems, we assess violations in the Lorenz gauge condition and find that the new method conserves charge at an acceptable level despite the fact that this is not strictly enforced.
PIC methods [2, 3] have been extensively applied in numerical simulations of kinetic plasmas and are an important class of techniques used in the design of experimental devices including lasers, pulsed power systems, particle accelerators, among others [4]. The earliest work involving these methods began in the 1950s and 1960s, and it remains an active area of research to this day. At its core, a PIC method combines an Eulerian approach for the electromagnetic fields with a Lagrangian method that evolves collections of samples taken from general distribution functions in phase space. In other words, the fields are evolved using a mesh, while the distribution function is evolved using particles whose equations of motion are set according to characteristics of the partial differential equations (PDEs) that govern the evolution of the distribution functions. Lastly, to combine the two approaches, an interpolation method is used to map data between the mesh and the particles. A typical selection for this map involves some combination of piece-wise constant or linear splines, with tensor products being used to address multi-dimensional problems.
The popularity of the PIC method in engineering applications can be largely attributed to its simplicity, efficiency, and capabilities in simulating complex nonlinear processes in plasmas. Early renditions of these methods were specifically constructed to circumvent the prohibitively expensive force calculation used to calculate the interactions between particles in electrostatic problems. The introduction of a mesh greatly simplified the force calculation because the number of mesh points is typically smaller than the number of particles and fast algorithms could be used to compute the potential. Simulation particles in these methods, which represent an ensemble of physical particles, are initialized by sampling from a prescribed probability distribution. A consequence of this sampling is that bulk processes in plasmas will be well represented, while the tails of the distribution will be largely underresolved even with good sampling methods. This, in turn, necessitates a large number of simulation particles for more systematic refinement studies to prevent certain numerical fluctuations. Realistic simulations of plasma devices have created a demand for new algorithms that simultaneously address the various challenges posed by accuracy constraints in modeling as well as scalability with new computational hardware [5]. The goal of this work is to supply new algorithms which aim to enhance the capabilities of existing PIC methods for plasmas. The prevalence of PIC, as a simulation tool, in the plasma physics community, has resulted in the introduction of numerous production codes with different capabilities. Despite these developments, however, comparisons that benchmark the performance of these algorithms have only recently been explored [6]. There is a clearly a need for more systematic benchmarking of existing PIC algorithms in the community, and we hope this work contributes significantly in this respect.
A comprehensive review of the literature for PIC methods up to 2005 can be found in the review article [7]. Much of the work highlighted by this reference is now largely considered standard, so, we shall emphasize more recent articles that are more aligned with the developments presented in this paper. Most PIC methods evolve the simulation particles explicitly using some form of leapfrog time integration along with the Boris rotation method [8] in the case of electromagnetic plasmas. The exploration of implicit PIC methods began in the 1980s [9, 10, 11]. These approaches suffered from a number of unattractive features, including issues with numerical heating and cooling [12], slow nonlinear convergence, and inconsistencies between the fluid moments and particle data. These approaches were later abandoned in favor of explicit treatments. Recent years have shown a resurgence of interest in implicit PIC methods [13, 14, 15]. In particular, the implicit PIC method proposed in [15] addressed many of these issues in the case of the Vlasov-Poisson system. Nonlinear convergence and self-consistency were enforced using a Jacobian-free Newton-Krylov method [16] with a novel fluid preconditioner [17] to enforce the continuity equation. This approach eliminated the need to resolve the charge separation in the plasma, which led to remarkable computational savings over explicit methods. These techniques were later extended to curved geometries through the use of smooth grid mappings [18]. Recently, an effort has been made to extend these techniques to the full Vlasov-Maxwell system [19] to avoid the highly restrictive CFL condition posed by the gyrofrequency, as well as the consideration of asymptotic-preserving treatments for particles [20, 21]. While these contributions are significant in their own right, there are many opportunities for improvement. Many of these methods are limited to second-order accuracy in both space and time and may greatly benefit from more accurate field solvers. Additionally, applications of interest involve complex geometries which introduce additional complications with stability and are often poorly resolved with uniform Cartesian meshes. Lastly, there is the concern of scalability. Krylov subspace methods present a considerable challenge for scalability on large machines due to the various collective operations used in the algorithms. For this reason, algorithms with explicit structure are more common in applications. It seems that the scalability of these methods could be significantly improved if similar implicit methods could be developed which eliminate these Krylov solves altogether, though this is beyond the scope of the present work.
A challenge associated with developing any solver for Maxwell’s equations is the enforcement of the involutions for the fields, namely ∇⋅E=ρ/ϵ0 and ∇⋅B=0. In the case of a structured Cartesian grid, Maxwell’s equations can be discretized using a staggered grid technique introduced by Yee [22]. The use of a staggered mesh yields a structure-preserving discrete analogue of Maxwell’s equations in integral form that automatically enforces the involutions for E and B without additional treatment. This is the basis of the well-known finite-difference time-domain method (FDTD) [23]. It is important to note that this is only true in the absence of moving charge. In the presence of moving charge, a standard linear current weighting will not satisfy the continuity equation ∂tρ+∇⋅J=0. In [1], maps for the current were constructed to properly ensure that Gauss’ law is satisfied. Divergence cleaning methods were also analyzed in [24] and were shown to be quite effective in simulations of a relativistic beam. Later, higher order versions of particle weighting were proposed in [25]. Then in [26], the authors generalized this approach to remove the assumption that particles move in straight lines. Such particle weightings can be useful for dense plasmas because they reduce numerical heating. However, they introduce complications in bounded domains, specifically when the plasma interacts with the boundary. In such cases, the approach of Villasenor and Buneman [1] is the preferred weighting for the current density. While the staggering in both space and time used in the original FDTD method is second-order accurate, a fourth-order extension of the spatial discretization was developed as a way of dealing with certain dispersion errors known as numerical Cerenkov radiation [27]. Pseudo-spectral type discretizations, free of numerical dispersion, have also been considered in simulations of relativistic plasmas [28, 29]. Recent work on relativistic plasmas has used both high-order weighting and high-order methods to study the long-term evolution in problems that require control of energy and momentum conservation [30]. In [31], a semi-implicit method was introduced to model systems in which the Debye length cannot be resolved for purposes of practicality. The method in [31] shares many of the same properties as the one presented in this work. While the use of a staggered mesh with finite-differences is quite effective for Cartesian grids, issues arise in problems defined with geometry, such as curved surfaces, in which one resorts to stair step boundaries [32]. To mitigate the effect of stair step boundaries in explicit methods, the mesh resolution is increased, resulting in a highly restrictive time step to meet the CFL stability criterion. Conformal PIC methods (see the review [33, 34]), which use smooth grid transformations to address curved boundaries have been developed to address some of these concerns, but there may be certain geometries for which (uniformly) small cells may be required to properly resolve the features. In the case of cut-cell meshes, the time step will be limited by the size of the smallest cell, which can be prohibitively restrictive. Another interesting approach for dealing with geometry in the Yee scheme, which avoids the stair stepping along the boundary, was developed for two-dimensional problems [35]. The grid cells along the boundary in the method are replaced with cut-cells that use generalized finite-difference updates to account for different intersections with the boundary. While this scheme was shown to be energy conserving, a more remarkable feature is that it eliminated the highly restrictive condition on the time step introduced by the cut-cells along the boundary. The theory in this article established half-order accuracy, yet demonstrated first-order accuracy in numerical experiments. While, these schemes have not yet been combined with PIC, they might eliminate some of the commonly encountered stability issues associated with cut-cells.
While many electromagnetic PIC methods solve Maxwell’s equations on Cartesian meshes through the FDTD method, other methods have been developed specifically for addressing issues posed by geometries through the use of unstructured meshes. In [36], a finite-element method (FEM) was coupled with PIC to model plasmas using the Darwin approximation, in which there is considerable (time) scale separation between the fields and the plasma. Truly conformal PIC methods based on the FEM have also been considered to address problems concerning general geometries and parallel scalability [37]. Explicit finite-volume methods (FVM), which can address geometry, were considered in [38], which also developed divergence cleaning methods suitable for applications to PIC simulations of the Vlasov-Maxwell system. Discontinuous Galerkin (DG) methods have also been used to develop high-order PIC methods with elliptic [39] and hyperbolic [40] divergence cleaning methods being employed to enforce Gauss’ law. Other work in this area has explored more generalized FEM discretizations in order to enforce charge conservation on arbitrary grids [41]. Structure-preserving discretizations [42, 43, 44, 45, 46, 47], which use exact sequence basis functions that follow the de Rham complex at a discrete level and automatically enforce involutions for the electric and magnetic fields, have also been proposed. Another method in this category was presented for the Vlasov-Maxwell system in [48], which exploited the Hamiltonian structure of the system to generate methods with numerous conservative properties that are independent of the basis functions. While the flexibility of these methods is quite appealing, such solvers rely on the solution of large systems of equations. Even with preconditioning, such methods can be slow and difficult implement in a scalable manner. In the case of explicit methods, such as FV and DG approaches, other challenges exist. The basic FVM, without additional reconstructions, is first-order accurate in space. These methods can, of course, be improved to second-order accuracy by performing reconstructions based on a collection of cells. Beyond second-order accuracy, the reconstruction process becomes quite complicated due to the growth in the size of the interpolation patches. DG methods, on the other hand, store cell-wise expansions in a basis, which eliminates the issue encountered in the FVM, typically at the cost of a highly restrictive condition on the size of a time step. Additionally, the significant amount of local work in DG methods makes them appealing for newer hardware, yet the restriction on the time step size is often left unaddressed. However, notable exceptions to this restriction exist for the two-way wave equation including staggered formulations [49] and Hermite methods [50], which allow for a much larger time step. It will be interesting to see the performance of such methods in plasma problems, especially in problems with intricate geometric features.
Other methods for Maxwell’s equations have been developed with unconditional stability for the time discretization. The first of these methods is the ADI-FDTD method [51, 52], which combined an ADI approach with a two-stage splitting to achieve an unconditionally stable solver. Time stepping in these methods was later generalized using a Crank-Nicolson splitting and several techniques for enhancing the temporal accuracy were proposed [53]. Of particular significance to this work are methods based on the method-of-lines-transpose (MOL^{T}) [54, 55, 56, 57]. These methods are unconditionally stable in time and can be obtained by reversing the typical order in which discretization is performed. By first discretizing in time, one can solve a resulting boundary-value problem by formally inverting a dimensionally-split differential operator using a Green’s function in conjunction with a fast summation method. Mesh-free methods for plasmas [58] have also been developed, which have been extended to Maxwell’s equations in the Darwin limit, under the Coulomb gauge [59, 60]. These formulations are in some ways similar to PIC in that they evolve particles with shapes with the exception that no mapping to a mesh is used in the simulation. The elliptic equations are solved using a Green’s function and a fast summation method is used for efficiency. Green’s function methods have also been used to develop asymptotic preserving schemes. In [61], a boundary integral formulation with a multi-dimensional Green’s function was used to construct a method that recovers the Darwin limit under appropriate conditions. The methods considered in the present work utilize dimensional splittings, which has been used to construct algorithms that are unconditionally stable, permit high-order time accuracy [56], show parallel scalability [62], and are geometrically flexible [63].
In [64], a PIC method was developed based on the MOLT discretization. This work leveraged a staggered grid formulation in which Maxwell’s equations were cast in terms of the Lorenz gauge. The wave equation for the scalar potential was replaced with an elliptic equation to control errors in the gauge condition. Since the particle equations were written in terms of E and B, additional finite-difference derivatives were required to compute the electric and magnetic fields from the potentials. In contrast, the formulation developed in this paper eliminates the use of a staggered mesh, so that data for the fields and particles are co-located. Furthermore, spatial derivatives of the potentials, which were originally computed using (low-order) finite-differences, are now evaluated analytically using the integral solution and converge at the same rate as the fields. In this work, we use quadrature that is fifth-order accurate in space. However, this approach naturally extends to arbitrary order and retains the geometric flexibility of the field solver. Lastly, the time integration method used for the particles in this work is fundamentally different from [64], which considered traditional leapfrog methods. The method we propose is based on a simple modification of a method presented in [60] and is compatible with the structure of the Hamiltonian formulation.
The contents of this paper are organized as follows. The relativistic formulation for the Vlasov-Maxwell system, which is the basis of the new PIC method developed in this work, is presented in § 2. Details concerning the treatment of the fields are presented in § 3, where we establish stability and time-consistency properties of the field solver at the semi-discrete level. In § 4, we describe the new PIC method that is based on the field solver of the previous section along with details of the time integration method. Section 5 presents time and space refinement results for the field solver along with an extensive collection of numerical results for the new PIC method. We conclude with a summary of the results and a discussion of future work in § 7.
Problem Formulation
In this section, we provide relevant details of the problem formulation used by the plasma applications considered in this work. We begin with a discussion of the relativistic Vlasov-Maxwell system, which is the most general model used in this work, in § 2.1. Then, once we have introduced the model, we discuss the treatment of the fields in § 2.2, which expresses Maxwell’s equations in terms of potentials. In this work, the fields are cast as wave equations through the Lorenz gauge condition. The generalized momentum formulation used for the particles is presented in § 2.3. We introduce components of the system in their dimensional form and provide their corresponding non-dimensionalizations in Appendix § 9, with the latter being used in the implementation. We then conclude the section with a brief summary to emphasize the key aspects of the proposed formulation.
Relativistic Vlasov-Maxwell System
In this work, we develop numerical algorithms for plasmas described by the relativistic Vlasov-Maxwell (VM) system, which in SI units, reads as
The first equation (1) is the relativistic Vlasov equation which describes the evolution of a probability distribution function fs(x,p,t) for particles of species s in phase space which have mass ms and charge qs. Here, we define γs=1/1+p2/(msc)2=1/1−v2/c2, which makes equation (1) Lorentz invariant. Physically, equation (1) describes the time evolution of a distribution function that represents the probability of finding a particle of species s at the position x, with linear momentum p=msγsv, at any given time t. Since the position and velocity data are vectors with 3 components, the distribution function is a scalar function of 6 dimensions plus time. While the equation itself has fairly simple structure, the primary challenge in numerically solving this equation is its high dimensionality. This growth in the dimensionality has posed tremendous difficulties for grid-based discretization methods, where one often needs to use many grid points to resolve relevant space and time scales in the problem. This difficulty is compounded by the fact that many plasmas of interest contain multiple species. Despite the lack of a collision operator on the right-hand side of (1), collisions occur in a mean-field sense through the electric and magnetic fields, which appear as coefficients of the gradient in momentum.
Equations (2) - (5) are Maxwell’s equations, which describe the evolution of the background electric and magnetic fields. Since the plasma is a collection of moving charges, any changes in the distribution function for each species will be reflected in the charge density ρ(x,t), as well as the current density J(x,t), which, respectively, are the source terms for Gauss’ law (4) and Ampère’s law (3). For Ns species, the total charge density and current density are defined by summing over the species
Here, the integrals are taken over the momentum components of phase space, which we have denoted by Ωp. The remaining parameters ϵ0 and μ0 are the permittivity and permeability of free-space. We also have the useful relation c2=1/(μ0ϵ0), where c denotes the speed of light. Equations (4) and (5) enforce charge conservation and prevent the appearance of so-called “magnetic monopoles.” It is imperative that numerical schemes for Maxwell’s equations satisfy these conditions. This is one of the reasons we adopt a gauge formulation for Maxwell’s equations, which is presented in the next section.
Maxwell's Equations with the Lorenz Gauge
Under the potential formulation, with the selection of the Lorenz gauge, Maxwell’s equations transform to a system of wave equations of the form
where c is the speed of light, ϵ0 and μ0 represent, respectively, the permittivity and permeability of free-space. Further, we have used ϕ to denote the scalar potential and A is the vector potential. In fact, under any choice of gauge condition, given ϕ and A, one can recover E and B via the relations
E=−∇ϕ−∂tA,B=∇×A,(8)
where “×” denotes the vector cross product. The structure of equations (8) and (9) is appealing because the system, modulo the gauge condition (10), is essentially a system of four “decoupled” scalar wave equations. Maxwell’s equations (2) - (5) are equivalent to (8) and (9) as long as the Lorenz gauge condition (10) is satisfied by ϕ and A. This formulation is appealing for several reasons. This form of the system is purely hyperbolic, so it evolves in a local sense. Computationally, this means that a localized method can be used to evolve the system, which will likely be more efficient for parallel computers. Another attractive feature is that many of the methods developed for scalar wave equations, e.g., [56, 57] can be applied to the system in a straightforward manner.
Hamiltonian Formulation for Relativistic Particles
To obtain the Hamiltonian formulation for the relativistic Vlasov-Maxwell system, we first introduce the Lagrangian for a single relativistic particle moving in a potential field, which can be shown to be
L=−γmc2−q(ϕ−A⋅v).(9)
Here, we have used m to denote the mass of the particle, q is its charge, v its velocity, c is the speed of light, γ=1/1−v2/c2 is the Lorentz factor, ϕ is the scalar potential, and A is the vector potential. Next, we define the generalized (conjugate) momentum from the relativistic Lagrangian (9) as
P=∂v∂L=γmv+qA.(10)
In addition, we have the following identities, which can be derived from (10):
v=mγ1(P−qA)=c2(P−qA)2+(mc2)2c2(P−qA).(11)
v2=v⋅v=(P−qA)2+(mc)2c2(P−qA)2.(12)
The Hamiltonian corresponding to the Lagrangian (9) can be obtained by means of a Legendre transform
H=P⋅v−L.(13)
Using the identities (11) and (12) in the transformation (13), we obtain the relativistic Hamiltonian
H=c2(P−qA)2+(mc2)2+qϕ.(14)
The Hamiltonian for a system of Np particles can be easily obtained by summing the Hamiltonians over individual particles, each having the form (14). From this, we can calculate the equations of motion for each particle using Hamilton’s equations, which leads to the system
References
[1]J. Villasenor and O. Buneman, “Rigorous charge conservation for local electromagnetic field solvers,” Computer Physics Communications, vol. 69, pp. 306–316, 2-3 1992.DOI
[2]C. K. Birdsall and A. B. Langdon, Plasma Physics via Computer Simulation, McGraw-Hill Book Company, 1985.
[3]R. W. Hockney and J. W. Eastwood, Computer Simulation Using Particles, First. CRC Press, 1988.
[4]B. B. Godfrey, “Review and recent advances in PIC modeling of relativistic beams and plasmas,” AIP Conference Proceedings, vol. 1777, 1 2016.
[5]S. Biedron et al., Snowmass21 accelerator modeling community white paper, 2022.arxiv.org/abs/2203.08335
[6]J. R. Smith, C. Orban, N. Rahman, B. McHugh, R. Oropeza, and E. A. Chowdhury, “A particle-in-cell code comparison for ion acceleration: EPOCH, LSP, and WarpX,” Physics of Plasmas, vol. 28, no. 7, p. 074505, Jul. 2021. DOI: 10.1063/5.0053109.DOI
[7]J. P. Verboncoeur, “Particle simulation of plasmas: Review and advances,” Plasma Physics and Controlled Fusion, vol. 47, A231–A260, 5A 2005.DOI
[8]Relativistic Plasma Simulation-Optimization of a Hybrid Code, (Naval Research Laboratory, Washington, D. C.), 1970, pp. 3–67.
[9]A. Langdon, B. Cohen, and A. Friedman, “Direct implicit large time-step particle simulation of plasmas,” Journal of Computational Physics, vol. 51, pp. 107–138, 1 1981.
[10]J. Brackbill and D. Forslund, “An implicit method for electromagnetic plasma simulation in two dimensions,” Journal of Computational Physics, vol. 46, pp. 271–308, 2 1982.DOI
[11]R. Mason, “An electromagnetic field algorithm for 2D implicit plasma simulation,” Journal of Computational Physics, vol. 71, pp. 429–473, 2 1987.DOI
[12]B. Cohen, A. Langdon, D. Hewett, and R. Procassini, “An implicit method for electromagnetic plasma simulation in two dimensions,” Journal of Computational Physics, vol. 81, pp. 151–168, 1 1989.
[13]G. Lapenta, J. Brackbill, and P. Ricci, “Kinetic approach to microscopic-macroscopic coupling in space and laboratory plasmas,” Physics of Plasmas, vol. 13, 5 2006.DOI
[14]S. Markidis and G. Lapenta, “The energy conserving particle-in-cell method,” Journal of Computational Physics, vol. 230, pp. 7037–7052, 18 2011.DOI
[15]G. Chen, L. Chacón, and D. Barnes, “An energy- and charge-conserving, implicit, electrostatic particle-in-cell algorithm,” Journal of Computational Physics, vol. 230, pp. 7018–7036, 18 2011.DOI
[16]D. Knoll and D. Keyes, “Jacobian-free Newton–Krylov methods: A survey of approaches and applications,” Journal of Computational Physics, vol. 193, pp. 357–397, 2 2004.DOI
[17]G. Chen, L. Chacón, C. Leibs, D. Knoll, and W. Taitano, “Fluid preconditioning for Newton–Krylov-based, fully implicit, electrostatic particle-in-cell simulations,” Journal of Computational Physics, vol. 258, pp. 555–567, 2014.DOI
[18]L. Chacón, G. Chen, and D. Barnes, “A charge-and energy-conserving implicit, electrostatic particle-in-cell algorithm on mapped computational meshes,” Journal of Computational Physics, vol. 233, pp. 1–9, 2012.DOI
[19]G. Chen, L. Chacón, L. Yin, B. Albright, J. Stark, and R. Bird, “A semi-implicit, energy- and charge-conserving particle-in-cell algorithm for the relativistic Vlasov-Maxwell equations,” Journal of Computational Physics, vol. 407, p. 109 228, 2020.DOI
[20]O. Koshkarov, L. Chacón, G. Chen, and L. F. Ricketson, “Fast nonlinear iterative solver for an implicit, energy-conserving, asymptotic-preserving charged-particle orbit integrator,” Journal of Computational Physics, vol. 459, 2022.DOI
[21]G. Chen and L. Chacón, An implicit, conservative and asymptotic-preserving electrostatic particle-in-cell algorithm for arbitrarily magnetized plasmas in uniform magnetic fields, 2022.DOI
[22]K. S. Yee, “Numerical solution of initial boundary value problems involving Maxwell’s equations in isotropic media,” IEEE Transactions on Antennas and Propagation, vol. 14, pp. 302–307, 3 1966.DOI
[23]A. Taflove and S. C. Hagness, Computational electrodynamics: the finite-difference time-domain method, Third. Artech House Publishers, 2005.
[24]P. Mardahl and J. Verboncoeur, “Charge conservation in electromagnetic PIC codes; spectral comparison of Boris/DADI and Langdon-Marder methods,” Computer Physics Communications, vol. 106, pp. 219–229, 1997.DOI
[25]T. Z. Esirkepov, “Exact charge conservation scheme for particle-in-cell simulation with an arbitrary form-factor,” Computer Physics Communications, vol. 135, no. 2, pp. 144–153, 2001.DOI
[26]T. Umeda, Y. Omura, T. Tominaga, and H. Matsumoto, “A new charge conservation method in electromagnetic particle-in-cell simulations,” Computer Physics Communications, vol. 156, no. 1, pp. 73–85, 2003.
[27]A. D. Greenwood, K. L. Cartwright, J. W. Luginsland, and E. A. Baca, “On the elimination of numerical Cerenkov radiation in PIC simulations,” Journal of Computational Physics, vol. 201, pp. 665–684, 2 2004.DOI
[28]J.-L. Vay, R. Lehe, H. Vincenti, B. Godfrey, I. Haber, and P. Lee, “Recent advances in high-performance modeling of plasma-based acceleration using the full PIC method,” Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment, vol. 829, pp. 353–357, 2016, 2nd European Advanced Accelerator Concepts Workshop - EAAC 2015, ISSN: 0168-9002. DOI: https://doi.org/10.1016/j.nima.2015.12.033.
[29]R. Lehe, M. Kirchen, I. A. Andriyash, B. B. Godfrey, and J.-L. Vay, “A spectral, quasi-cylindrical and dispersion-free particle-in-cell algorithm,” Computer Physics Communications, vol. 203, pp. 66–82, 2016, ISSN: 0010-4655. DOI: https://doi.org/10.1016/j.cpc.2016.02.007.
[30]M. Shalaby, A. E. Broderick, P. Chang, C. Pfrommer, A. Lamberts, and E. Puchwein, “Sharp: A spatially higher-order, relativistic particle-in-cell code,” The Astrophysical Journal, vol. 841, no. 1, p. 52, 2017.DOI
[31]F. Bacchini, “Relsim: A relativistic semi-implicit method for particle-in-cell simulations,” arXiv preprint arXiv:2306.04685, 2023.arxiv.org/abs/2306.04685
[32]J. P. Verboncoeur, “Aliasing of electromagnetic fields in stair step boundaries,” Computer physics communications, vol. 164, pp. 344–352, 1 2004.DOI
[33]C. S. Meierbachtol, A. D. Greenwood, J. P. Verboncoeur, and B. Shanker, “Conformal electromagnetic particle in cell: A review,” IEEE Transactions on Plasma Science, vol. 43, 11 2015.DOI
[34]Y. Wang, J. Wang, Z. Chen, G. Cheng, and P. Wang, “Three-dimensional simple conformal symplectic particle-in-cell methods for simulations of high power microwave devices,” Computer Physics Communications, vol. 205, pp. 1–12, 2016.DOI
[35]B. Engquist, J. Häggblad, and O. Runborg, “On energy preserving consistent boundary conditions for the Yee scheme in 2D,” BIT Numerical Mathematics, vol. 52, pp. 615–637, 3 2012.DOI
[36]E. Sonnendrücker, J. Ambrosiano, and S. Brandon, “A finite element formulation of the Darwin PIC model for use on unstructured grids,” Journal of Computational Physics, vol. 121, pp. 281–297, 2 1995.DOI
[37]J. Eastwood, W. Arter, N. Brealey, and R. Hockney, “Body-fitted electromagnetic pic software for use on parallel computers,” Computer Physics Communications, vol. 87, pp. 155–178, 1-2 1995.DOI
[38]C.-D. Munz, P. Ommes, R. Schneider, E. Sonnendrücker, and U. Voss, “Divergence correction techniques for Maxwell solvers based on a hyperbolic model,” Journal of Computational Physics, vol. 161, no. 2, pp. 484–511, 2000.DOI
[39]G. Jacobs and J. Hesthaven, “High-order nodal discontinuous Galerkin particle-in-cell method on unstructured grids,” Journal of Computational Physics, vol. 214, pp. 96–121, 1 2006.DOI
[40]G. Jacobs and J. Hesthaven, “Implicit–explicit time integration of a high-order particle-in-cell method with hyperbolic divergence cleaning,” Journal of Computational Physics, vol. 180, pp. 1760–1767, 10 2009.DOI
[41]M. Pinto, S. Jund, S. Salmon, and E. Sonnendrücker, “Charge-conserving FEM–PIC schemes on general grids,” Comptes Rendus Mécanique, vol. 342, pp. 570–582, 10-11 2014.DOI
[42]M. Pinto, K. Kormann, and E. Sonnendrücker, Variational framework for structure-preserving electromagnetic particle-in-cell methods, 2021. DOI: 10.48550/ARXIV.2101.09247. [Online]. Available: https://arxiv.org/abs/2101.09247.
[43]S. O’Connor, Z. D. Crawford, J. P. Verboncoeur, J. Luginsland, and B. Shanker, “A set of benchmark tests for validation of 3-D particle in cell methods,” IEEE Transactions on Plasma Science, vol. 49, pp. 1724–1731, 5 Apr. 2021.
[44]S. O’Connor, Z. D. Crawford, O. H. Ramachandran, J. Luginsland, and B. Shanker, “Time integrator agnostic charge conserving finite element PIC,” Physics of Plasmas, vol. 28, p. 092111, 9 Sep. 2021.DOI
[45]Z. D. Crawford, S. O’Connor, J. Luginsland, and B. Shanker, “Rubrics for charge conserving current mapping in finite element electromagnetic particle in cell methods,” IEEE Transactions on Plasma Science, vol. 49, pp. 3719–3732, 11 Nov. 2021.DOI
[46]M. Campos Pinto and V. Pagès, “A semi-implicit electromagnetic FEM-PIC scheme with exact energy and charge conservation,” Journal of Computational Physics, vol. 453, p. 110912, 2022. DOI: 10.1016/j.jcp.2021.110912.
[47]K. Kormann and E. Sonnendrücker, “Energy-conserving time propagation for a structure-preserving particle-in-cell Vlasov–Maxwell solver,” Journal of Computational Physics, vol. 425, p. 109890, 15 2021. DOI: 10.1016/j.jcp.2020.109890.DOI
[48]M. Kraus, K. Kormann, P. Morrison, and E. Sonnendrücker, “GEMPIC: Geometric electromagnetic particle-in-cell methods,” Journal of Plasma Physics, vol. 83, 4 2017. DOI: 10.1017/S002237781700040X.DOI
[49]D. Appelö, L. Zhang, T. Hagstrom, and F. Li, “An energy-based discontinuous Galerkin method with tame CFL numbers for the wave equation,” 2021. DOI: 10.48550/ARXIV.2110.07099. [Online]. Available: https://arxiv.org/abs/2110.07099.
[50]O. Beznosov and D. Appelö, “Hermite-discontinuous Galerkin overset grid methods for the scalar wave equation,” Communications on Applied Mathematics and Computation, vol. 3, pp. 391–418, 3 2021.arxiv.org/abs/2001.05072
[51]F. Zheng, Z. Chen, and J. Zhang, “A finite-difference time-domain method without the Courant stability conditions,” IEEE Microwave and Guided Wave Letters, vol. 9, pp. 497–523, 11 1999.DOI
[52]F. Zheng, Z. Chen, and J. Zhang, “Toward the development of a three-dimensional unconditionally stable finite-difference time-domain method,” IEEE Transactions on Microwave Theory and Techniques, vol. 48, pp. 1550–1558, 9 2000.DOI
[53]J. Lee and B. Fornberg, “Some unconditionally stable time stepping methods for the 3D Maxwell’s equations,” Journal of Computational and Applied Mathematics, vol. 166, pp. 497–523, 2 2004.DOI
[54]M. F. Causley, A. J. Christlieb, Y. Güçlü, and E. Wolf, “Method of lines transpose: A fast implicit wave propagator,” 2013. DOI: 10.48550/ARXIV.1306.6902. [Online]. Available: https://arxiv.org/abs/1306.6902.
[55]M. Causley, A. Christlieb, B. Ong, and L. Van Groningen, “Method of lines transpose: An implicit solution to the wave equation,” Mathematics of Computation, vol. 83, no. 290, pp. 2763–2786, 2014.DOI
[56]M. F. Causley and A. J. Christlieb, “Higher order A-stable schemes for the wave equation using a successive convolution approach,” SIAM Journal on Numerical Analysis, vol. 52, no. 1, pp. 220–235, 2014.
[57]M. Causley, A. Christlieb, and E. Wolf, “Method of lines transpose: An efficient unconditionally stable solver for wave propagation,” Journal of Scientific Computing, vol. 70, no. 2, pp. 896–921, 2017.
[58]A. Christlieb, R. Krasny, J. Verboncoeur, J. Emhoff, and I. Boyd, “Grid-free plasma simulation techniques,” IEEE Transactions on Plasma Science, vol. 34, pp. 149–165, 2 2006.DOI
[59]M. Mäusek and P. Gibbon, “Mesh-free magnetoinductive plasma model,” IEEE Transactions on Plasma Science, vol. 38, pp. 2377–2382, 9 2010.DOI
[60]L. Siddi, G. Lapenta, and P. Gibbon, “Mesh-free Hamiltonian implementation of two dimensional Darwin model,” Physics of Plasmas, vol. 24, pp. 1–11, 8 2017.DOI
[61]Y. Cheng, A. J. Christlieb, W. Guo, and B. Ong, “An asymptotic preserving Maxwell solver resulting in the Darwin limit of electrodynamics,” Journal of Scientific Computing, vol. 71, no. 3, pp. 959–993, 2017.arxiv.org/abs/1606.08867
[62]A. J. Christlieb, P. T. Guthrey, W. A. Sands, and M. Thavappiragasm, “Parallel algorithms for successive convolution,” Journal of Scientific Computing, vol. 86, pp. 1–44, 1 2021.
[63]M. Thavappiragasm, A. Christlieb, J. Luginsland, and P. Guthrey, “A fast local embedded boundary method suitable for high power electromagnetic sources,” AIP Advances, vol. 10, p. 115318, 11 2020.DOI
[64]E. M. Wolf, M. Causley, A. Christlieb, and M. Bettencourt, “A particle-in-cell method for the simulation of plasmas based on an unconditionally stable field solver,” Journal of Computational Physics, vol. 326, pp. 342–372, 2016.
[65]A. J. Christlieb, R. Krasny, and J. P. Verboncoeur, “A treecode algorithm for simulating electron dynamics in a Penning–Malmberg trap,” Computer physics communications, vol. 164, pp. 306–310, 1-3 2004.DOI
[66]A. J. Christlieb, R. Krasny, and J. P. Verboncoeur, “Efficient particle simulation of a virtual cathode using a grid-free treecode poisson solver,” IEEE Transactions on Plasma Science, vol. 32, pp. 384–389, 2 2004.
[67]S. K. Veerapaneni, D. Gueyffier, D. Zorin, and G. Biros, “A boundary integral method for simulating the dynamics of inextensible vesicles suspended in a viscous fluid in 2D,” Journal of Computational Physics, vol. 228, pp. 2334–2353, 7 2009.DOI
[68]F. Vico, M. Ferrando, L. Greengard, and Z. Gimbutas, “The decoupled potential integral equation for time-harmonic electromagnetic scattering,” Communications on Pure and Applied Mathematics, vol. 69, pp. 771–812, 4 2016.
[69]A. Barnett, L. Greengard, and T. Hagstrom, “High-order discretization of a stable time-domain integral equation for 3D acoustic scattering,” Journal of Computational Physics, vol. 402, p. 109 047, 2020.
[70]S.-H. Chiu, M. N. J. Moore, and B. Quaife, “Viscous transport in eroding porous media,” Journal of Fluid Mechanics, vol. 893, A3, 2020. DOI: 10.1017/jfm.2020.228.DOI
[71]M. C. A. Kropinski and B. D. Quaife, “Fast integral equation methods for Rothe’s method applied to the isotropic heat equation,” Computers and Mathematics with Applications, vol. 61, pp. 2436–2446, 9 2011.
[72]M. Thavappiragasm, A. Viswanathan, and A. Christlieb, “MOLT based fast high-order three dimensional A-stable scheme for wave propagation,” Journal of Coupled Systems and Multiscale Dynamics, vol. 5, no. 2, pp. 151–163, 2017.
[73]G. B. Folland, Introduction to Partial Differential Equations: Second edition. Princeton University Press, 1995, ISBN: 978-0691213033.
[74]D. G. Duffy, Green’s Functions with Applications. CRC Press, 2001, ISBN: 978-1584881100.
[75]M. Causley, H. Cho, and A. Christlieb, “Method of lines transpose: Energy gradient flows using direct operator inversion for phase-field models,” SIAM Journal on Scientific Computing, vol. 39, no. 5, B968–B992, 2017.
[76]A. Christlieb, W. Sands, and H. Yang, “A kernel-based explicit unconditionally stable scheme for Hamilton-Jacobi equations on nonuniform meshes,” Journal of Computational Physics, vol. 415, pp. 1–25, 2020, Art. No. 109543.
[77]B. Leimkuhler and S. Reich, Simulating Hamiltonian Dynamics (Cambridge Monographs on Applied and Computational Mathematics). Cambridge University Press, 2005. DOI: 10.1017/CBO9780511614118.DOI
[78]M. Tao, “Explicit symplectic approximation of nonseparable Hamiltonians: Algorithm and long time performance,” Phys. Rev. E, vol. 94, p. 043303, 4 Oct. 2016. DOI: 10.1103/PhysRevE.94.043303.DOI
[79]E. Wolf, “A particle-in-cell method for the simulation of plasmas based on an unconditionally stable wave equation solver,” Ph.D. dissertation, Michigan State University, 2015.
[80]R. Liboff, Kinetic Theory: Classical, Quantum, and Relativistic Descriptions. Springer-Verlag New York, 2003, ISBN: 978-0-387-95551-3.
[82]J. A. Bittencourt, Fundamentals of Plasma Physics, Third. Springer-Verlag, 2010.
[83]I. Langmuir, “Electric discharges in gases at low pressures,” Journal of the Franklin Institute, vol. 214, no. 3, pp. 275–298, 1932, ISSN: 0016-0032. DOI: https://doi.org/10.1016/S0016-0032(32)90932-6. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0016003232909326.
[84]M. A. Lieberman and A. J. Lichtenberg, Principles of Plasma Discharges and Materials Processing. New York: Wiley-Interscience, 1994, ISBN: 978-0471005773.
[85]J. Allen, “The plasma–sheath boundary: Its history and Langmuir’s definition of the sheath edge,” Plasma Sources Science and Technology, vol. 18, no. 1, p. 014004, Nov. 2008. DOI: 10.1088/0963-0252/18/1/014004. [Online]. Available: https://dx.doi.org/10.1088/0963-0252/18/1/014004.
[86]N. Vaughn, L. Wilson, and R. Krasny, “A GPU-accelerated barycentric Lagrange trecode,” in 2020 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW), 2020, pp. 701–710. DOI: 10.1109/IPDPSW50202.2020.00125.DOI
[87]L. Wilson, N. Vaughn, and R. Krasny, “A GPU-accelerated fast multipole method based on barycentric Lagrange interpolation and dual tree traversal,” Computer Physics Communications, vol. 265, 2021, ISSN: 0010-4655. DOI: https://doi.org/10.1016/j.cpc.2021.108017. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S0010465521001296.
[88]A. Christlieb, W. Guo, Y. Jiang, and H. Yang, “Kernel based high order explicit unconditionally-stable scheme for nonlinear degenerate advection-diffusion equations,” Journal of Scientific Computing, vol. 82:52, pp. 1–29, 2020.
[89]G.-S. Jiang and C.-W. Shu, “Efficient implementation of weighted ENO schemes,” Journal of computational physics, vol. 126, no. 1, pp. 202–228, 1996.
[90]A. Christlieb, W. Guo, and Y. Jiang, “A WENO-based method of lines transpose approach for Vlasov simulations,” Journal of Computational Physics, vol. 327, pp. 337–367, 2016.
[91]H. Kreiss, N. Petersson, and J. Yström, “Difference approximations of the Neumann problem for the second order wave equation,” SIAM Journal of Numerical Analysis, vol. 42, pp. 1292–1323, 3 2004.