pith. sign in

arxiv: 2301.01753 · v2 · submitted 2023-01-04 · 🧮 math.NA · cs.NA· physics.comp-ph

Generalized Yee methods: Scalable symplectic finite element Maxwell solvers

Alexander S. Glasser , Hong Qin This is my paper

Pith reviewed 2026-05-24 09:59 UTC · model grok-4.3

classification 🧮 math.NA cs.NAphysics.comp-ph
keywords finite element methodsMaxwell equationssymplectic methodssparse approximate inversesYee methodparticle-in-cellstructure preservationnumerical analysis
0
0 comments X

The pith

Generalized Yee methods extend the locality and symplecticity of Yee's scheme to finite-element Maxwell solvers on unstructured meshes.

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

The paper shows that generalized Yee methods, constructed from de Rham-conforming finite elements and sparse approximate inverses, preserve both the locality and the symplectic structure of Maxwell's equations. This preservation enables scalability on high-performance computers and long-time numerical accuracy, just as in the original Yee method. The key result is that symplecticity remains invariant under sparse approximations, allowing flexibility in how the mass matrices are sparsified. The methods also support higher-order elements and particle-in-cell coupling for electromagnetic simulations while maintaining structure preservation.

Core claim

Generalized Yee methods are built from de Rham-conforming finite elements that achieve locality through sparse mass matrices and their sparse approximate inverses, with a proof that the symplectic structure is invariant under such sparse approximations, thereby generalizing Yee's method to unstructured meshes, higher-order accuracy, and symplectic particle-in-cell methods.

What carries the argument

de Rham-conforming finite elements combined with sparse approximate inverses that preserve the discrete symplectic form of Maxwell's equations.

Load-bearing premise

De Rham-conforming finite elements on unstructured meshes can be paired with sparse approximate inverses while retaining both locality and the required discrete symplectic form.

What would settle it

Demonstrating that the discrete symplectic form is not preserved, or that long-time energy drift occurs, when applying a GYM to a standard Maxwell test problem on an unstructured mesh.

Figures

Figures reproduced from arXiv: 2301.01753 by Alexander S. Glasser, Hong Qin.

Figure 1
Figure 1. Figure 1: The generalized Whitney 1-forms Wx1x2 and Wx1x2x4x3 are schematically depicted, respectively, on the left and right. Wx1x2 evaluates to the length |x1x2| when integrated along the edge x1x2 (in blue) and vanishes on all other edges (in orange). Wx1x2x4x3 likewise yields the area |x1x2x4x3| when integrated over the blue face x1x2x4x3, and vanishes on all other faces. 5 [PITH_FULL_IMAGE:figures/full_fig_p00… view at source ↗
Figure 2
Figure 2. Figure 2: A depiction of the degrees of freedom in￾volved in C T—the transposed curl operator—for gener￾alized Whitney forms on a cubic mesh. simplicity, we use Eq. (3) to integrate inner products of forms such as those appearing in Eq. (18) to find that (M1)σ1,τ1 = ∆V ·    4/9 σ 1 = τ 1 1/9 σ 1 k τ 1 and ∃ τ 2 ⊃ {σ 1 , τ 1} 1/36 σ 1 k τ 1 and ∃ τ 3 ⊃ {σ 1 , τ 1} (M2)σ2,τ2 = ∆V · ( 2/3 σ 2 = τ 2 1/6 σ 2 k τ 2 a… view at source ↗
Figure 3
Figure 3. Figure 3: The figures above depict the L 2Λ 1 log relative error in an FEEC approximation of E˙ = c 2∇ × ∇ × A vs. log cell size h. The x-axis measures the number of cells (of size h) per wavelength λA = 2π/kn of the vector potential A = sin(kny)dx. The left plot uses the first order (Whitney form) P − 1 Λ p (Th) FEEC basis, while the right plot uses a more accurate second order basis, P − 2 Λ p (Th). For each basis… view at source ↗
read the original abstract

Yee's finite-difference method preserves two crucial properties of Maxwell's equations -- locality and symplecticity -- and thereby enjoys two computational advantages: scalability on high-performance architectures and long-time numerical accuracy. In this work, we show that Yee's method is a special case of a class of structure-preserving finite element methods -- termed generalized Yee methods (GYMs) -- that are designed to retain both crucial properties. GYMs are built from de Rham-conforming finite elements and achieve locality through sparse mass matrices and their sparse approximate inverses (SPAIs). We prove that the symplectic structure of GYMs is invariant under such sparse approximations, freeing the choice of sparsification strategy. We introduce a novel sparsification strategy, SPAI-OP, which concentrates accuracy at prescribed wave modes by operator probing. We further extend GYMs to structure-preserving electromagnetic particle-in-cell (PIC) methods, whose symplecticity over particle trajectories requires the smooth fields afforded by higher-order finite elements. GYMs therefore retain the computational virtues of Yee's method while enabling unstructured meshes, higher-order accuracy, spectral adaptivity, and symplectic particle coupling.

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 introduces Generalized Yee Methods (GYMs) as a class of de Rham-conforming finite-element discretizations of Maxwell's equations that generalize Yee's FDTD scheme while retaining locality (via sparse mass matrices and SPAIs) and symplecticity. The central claim is a proof that the discrete symplectic structure remains invariant under replacement of the exact inverse mass matrix by any sparse approximate inverse, thereby freeing the choice of sparsification; the work also presents the SPAI-OP strategy and extends GYMs to structure-preserving electromagnetic PIC methods.

Significance. If the invariance result holds without hidden symmetry or orthogonality restrictions on the SPAI, the paper would supply a scalable, long-time-accurate route to unstructured-mesh, higher-order Maxwell solvers that inherit the computational virtues of Yee's method. The explicit construction of SPAI-OP and the PIC extension are concrete contributions that could be adopted independently of the invariance theorem.

major comments (2)
  1. [Proof of symplectic invariance (likely §3–4)] The proof that symplectic invariance holds for arbitrary SPAIs (the load-bearing claim freeing all sparsification strategies) must be checked against the algebraic requirement that the resulting discrete operator remain skew-symmetric with respect to the de Rham pairing. On unstructured meshes a generic SPAI need not preserve symmetry of the mass-matrix inverse; if the argument tacitly assumes a symmetric approximant or an additional orthogonality condition, the statement in the abstract does not follow for the full advertised class of SPAIs.
  2. [§5, electromagnetic PIC extension] The extension to symplectic PIC methods (§5) relies on the smooth fields produced by higher-order elements to guarantee symplecticity along particle trajectories. The manuscript should supply an explicit statement of the discrete symplectic form for the coupled system and verify that the SPAI replacement does not destroy the required cancellation when the particle current is projected onto the same de Rham complex.
minor comments (2)
  1. [SPAI-OP construction] Notation for the SPAI-OP operator (Eq. (??)) should be introduced with a short algorithmic box or pseudocode so that the probing step is reproducible without reference to external literature.
  2. [Numerical results section] Figure captions for the unstructured-mesh examples should state the polynomial degree, mesh size, and time-step size used, together with the observed long-time energy drift (or lack thereof) to make the numerical evidence directly comparable to the theoretical claim.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful reading and constructive comments on the proof of symplectic invariance and the PIC extension. We respond point-by-point below.

read point-by-point responses

  1. Referee: [Proof of symplectic invariance (likely §3–4)] The proof that symplectic invariance holds for arbitrary SPAIs (the load-bearing claim freeing all sparsification strategies) must be checked against the algebraic requirement that the resulting discrete operator remain skew-symmetric with respect to the de Rham pairing. On unstructured meshes a generic SPAI need not preserve symmetry of the mass-matrix inverse; if the argument tacitly assumes a symmetric approximant or an additional orthogonality condition, the statement in the abstract does not follow for the full advertised class of SPAIs.

    Authors: The proof in §§3–4 establishes invariance for arbitrary SPAIs without assuming symmetry of the approximant or extra orthogonality. The discrete symplectic form is defined directly via the de Rham pairing; replacing the exact inverse mass matrix by any SPAI preserves skew-symmetry because the same approximant appears consistently on both sides of the variational equations, so the required cancellations hold by algebraic identity independent of mesh structure. We will add a short clarifying remark after the main theorem to make this explicit. revision: partial

  2. Referee: [§5, electromagnetic PIC extension] The extension to symplectic PIC methods (§5) relies on the smooth fields produced by higher-order elements to guarantee symplecticity along particle trajectories. The manuscript should supply an explicit statement of the discrete symplectic form for the coupled system and verify that the SPAI replacement does not destroy the required cancellation when the particle current is projected onto the same de Rham complex.

    Authors: We agree an explicit statement strengthens the section. In the revision we will insert the definition of the coupled discrete symplectic form (the sum of the field symplectic form and the particle canonical form) and verify that the SPAI acts only on the field equations after the current has been projected onto the de Rham complex; the projection commutes with the discrete exterior derivative, so the cancellation that yields symplecticity along trajectories is unaffected by the mass-matrix approximation. Higher-order elements supply the required smoothness for the particle push. revision: yes

Circularity Check

0 steps flagged

Minor self-citation not load-bearing; invariance proof presented as independent

full rationale

The paper's central claim is a proof that symplectic structure is invariant under sparse approximations of the mass matrix inverse. No equations or steps in the provided abstract reduce the result to a fitted parameter, self-definition, or renamed empirical pattern. Any self-citations (common in structure-preserving FEM literature) do not appear to carry the load of the invariance statement, which is framed as a direct algebraic result on de Rham elements. The derivation chain therefore remains self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The central claims rest on the existence of de Rham-conforming elements that discretely preserve the exterior derivative and on the algebraic property that the symplectic form survives replacement of the mass matrix by any sparse approximate inverse.

axioms (2)
  • domain assumption de Rham-conforming finite elements discretely preserve the differential structure of Maxwell's equations
    Invoked to build the GYM class from standard finite-element spaces.
  • ad hoc to paper the discrete symplectic form remains invariant when the mass matrix is replaced by any sparse approximate inverse
    This is the load-bearing statement proved in the paper; it is not a standard result in the finite-element literature referenced by the abstract.

pith-pipeline@v0.9.0 · 5728 in / 1342 out tokens · 18476 ms · 2026-05-24T09:59:29.615842+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

41 extracted references · 41 canonical work pages · 1 internal anchor

  1. [1]

    Numerical solution of initial boundary value problems involving maxwell’s equations in isotropic media,

    K. Yee, “Numerical solution of initial boundary value problems involving maxwell’s equations in isotropic media,” IEEE Transactions on Anten- nas and Propagation, vol. 14, no. 3, pp. 302–307, 1966

    work page 1966

  2. [2]

    Taflove and S

    A. Taflove and S. C. Hagness, Computa- tional electrodynamics: the finite-difference time-domain method, 3rd ed. Boston: Artech House, 2005

    work page 2005

  3. [3]

    Hairer, C

    E. Hairer, C. Lubich, and G. Wanner, Geo- metric Numerical Integration , 2nd ed. Berlin: Springer, 2006

    work page 2006

  4. [4]

    Whitney, Geometric Integration Theory

    H. Whitney, Geometric Integration Theory . Princeton, NJ: Princeton University Press, 1957

    work page 1957

  5. [5]

    Discrete Exterior Calculus

    M. Desbrun, A. N. Hirani, M. Leok, and J. E. Marsden, “Discrete exterior calculus,” arXiv preprint math/0508341 , 2005. [Online]. Available: https://arxiv.org/abs/math/0508341

    work page internal anchor Pith review Pith/arXiv arXiv 2005

  6. [6]

    Finite element exterior calculus, homological techniques, and applications,

    D. N. Arnold, R. S. Falk, and R. Winther, “Finite element exterior calculus, homological techniques, and applications,” Acta Numerica, vol. 15, 2006

    work page 2006

  7. [7]

    Finite element exterior calculus: from Hodge theory to numerical stability,

    ——, “Finite element exterior calculus: from Hodge theory to numerical stability,” Bulletin of the American Mathematical Society , vol. 47, no. 2, pp. 281–354, 2010

    work page 2010

  8. [8]

    Symplectic integrators and their application to dynamical astronomy,

    H. Kinoshita, H. Yoshida, and H. Nakai, “Symplectic integrators and their application to dynamical astronomy,” CELESTIAL ME- CHANICS AND DYNAMICAL ASTRONOMY, vol. 50, no. 1, pp. 59–71, 1991

    work page 1991

  9. [9]

    Sym- plectic integrators for long-term integrations in celestial mechanics,

    B. Gladman, M. Duncan, and J. Candy, “Sym- plectic integrators for long-term integrations in celestial mechanics,” Celestial Mechanics and Dynamical Astronomy, vol. 52, no. 3, pp. 221– 240, 1991

    work page 1991

  10. [10]

    Symplectic Integrator Algo- rithms for Modeling Planetary Accretion in Bi- nary Star Systems,

    J. E. Chambers, E. V. Quintana, M. J. Duncan, and J. J. Lissauer, “Symplectic Integrator Algo- rithms for Modeling Planetary Accretion in Bi- nary Star Systems,” The Astronomical Journal, vol. 123, no. 5, pp. 2884–2894, 2002

    work page 2002

  11. [11]

    Numerical integration in celestial me- chanics: a case for contact geometry,

    A. Bravetti, M. Seri, M. Vermeeren, and F. Zadra, “Numerical integration in celestial me- chanics: a case for contact geometry,” Celestial Mechanics and Dynamical Astronomy , vol. 132, no. 1, 2020

    work page 2020

  12. [12]

    Discrete gravity with local Lorentz invariance,

    E. Kur and A. S. Glasser, “Discrete gravity with local Lorentz invariance,” Physical Review D , vol. 106, no. 6, p. 064001, 2022

    work page 2022

  13. [13]

    Structure-preserving modelling of elastic waves,

    X. Li, W. Wang, M. Lu, M. Zhang, and Y. Li, “Structure-preserving modelling of elastic waves,” Geophysical Journal International , vol. 188, no. 3, pp. 1382–1392, 2012

    work page 2012

  14. [14]

    A modified symplectic scheme for seismic wave modeling,

    S. Liu, X. Li, W. Wang, L. Xu, and B. Li, “A modified symplectic scheme for seismic wave modeling,” Journal of Applied Geophysics , vol. 116, pp. 110–120, 2015. 12

    work page 2015

  15. [15]

    Geometric integration of the Vlasov-Maxwell system with a variational particle-in-cell scheme,

    J. Squire, H. Qin, and W. M. Tang, “Geometric integration of the Vlasov-Maxwell system with a variational particle-in-cell scheme,” Physics of Plasmas, vol. 19, no. 8, p. 084501, 2012

    work page 2012

  16. [16]

    Explicit high-order non-canonical symplectic particle-in-cell algorithms for Vlasov- Maxwell systems,

    J. Xiao, H. Qin, J. Liu, Y. He, R. Zhang, and Y. Sun, “Explicit high-order non-canonical symplectic particle-in-cell algorithms for Vlasov- Maxwell systems,” Physics of Plasmas , vol. 22, no. 11, p. 112504, 2015

    work page 2015

  17. [17]

    Hamiltonian time integrators for Vlasov-Maxwell equations,

    Y. He, H. Qin, Y. Sun, J. Xiao, R. Zhang, and J. Liu, “Hamiltonian time integrators for Vlasov-Maxwell equations,” Physics of Plasmas, vol. 22, no. 12, p. 124503, 2015

    work page 2015

  18. [18]

    Hamiltonian splitting for the Vlasov–Maxwell equations,

    N. Crouseilles, L. Einkemmer, and E. Faou, “Hamiltonian splitting for the Vlasov–Maxwell equations,” Journal of Computational Physics , vol. 283, pp. 224–240, feb 2015

    work page 2015

  19. [19]

    Canonical symplectic particle-in-cell method for long-term large-scale simulations of the Vlasov–Maxwell equations,

    H. Qin, J. Liu, J. Xiao, R. Zhang, Y. He, Y. Wang, Y. Sun, J. W. Burby, L. Ellison, and Y. Zhou, “Canonical symplectic particle-in-cell method for long-term large-scale simulations of the Vlasov–Maxwell equations,” Nuclear Fusion, vol. 56, no. 1, p. 014001, 2016

    work page 2016

  20. [20]

    GEMPIC: Geometric Elec- troMagnetic Particle-In-Cell Methods,

    M. Kraus, K. Kormann, P. J. Morrison, and E. Sonnendr¨ ucker, “GEMPIC: Geometric Elec- troMagnetic Particle-In-Cell Methods,” Journal of Plasma Physics , vol. 83, no. 4, 2017

    work page 2017

  21. [21]

    Structure and structure- preserving algorithms for plasma physics,

    P. J. Morrison, “Structure and structure- preserving algorithms for plasma physics,” Physics of Plasmas , vol. 24, no. 5, p. 055502, 2017

    work page 2017

  22. [22]

    The geometric theory of charge conservation in particle-in-cell simula- tions,

    A. S. Glasser and H. Qin, “The geometric theory of charge conservation in particle-in-cell simula- tions,” Journal of Plasma Physics, vol. 86, no. 3, p. 835860303, 2020

    work page 2020

  23. [23]

    A gauge-compatible Hamiltonian split- ting algorithm for particle-in-cell simulations us- ing finite element exterior calculus,

    ——, “A gauge-compatible Hamiltonian split- ting algorithm for particle-in-cell simulations us- ing finite element exterior calculus,” Journal of Plasma Physics , vol. 88, no. 2, p. 835880202, 2022

    work page 2022

  24. [24]

    Geometric Discretization of La- grangian Mechanics and Field Theories,

    A. Stern, “Geometric Discretization of La- grangian Mechanics and Field Theories,” Ph.D. dissertation, California Institute of Technology, Pasadena, California, 2009

    work page 2009

  25. [25]

    Sparse and explicit FETD via approximate inverse Hodge (mass) matrix,

    Bo He and F. Teixeira, “Sparse and explicit FETD via approximate inverse Hodge (mass) matrix,” IEEE Microwave and Wireless Com- ponents Letters, vol. 16, no. 6, pp. 348–350, Jun. 2006

    work page 2006

  26. [26]

    Differential Forms, Galerkin Duality, and Sparse Inverse Approxi- mations in Finite Element Solutions of Maxwell Equations,

    B. He and F. L. Teixeira, “Differential Forms, Galerkin Duality, and Sparse Inverse Approxi- mations in Finite Element Solutions of Maxwell Equations,” IEEE Transactions on Antennas and Propagation, vol. 55, no. 5, pp. 1359–1368, May 2007

    work page 2007

  27. [27]

    Parallel and Ex- plicit Finite-Element Time-Domain Method for Maxwell’s Equations,

    J. Kim and F. L. Teixeira, “Parallel and Ex- plicit Finite-Element Time-Domain Method for Maxwell’s Equations,” IEEE Transactions on Antennas and Propagation , vol. 59, no. 6, pp. 2350–2356, Jun. 2011

    work page 2011

  28. [28]

    Differential Forms in Lattice Field Theories: An Overview,

    F. L. Teixeira, “Differential Forms in Lattice Field Theories: An Overview,” ISRN Mathemat- ical Physics, vol. 2013, pp. 1–16, Feb. 2013

    work page 2013

  29. [29]

    Generalized time-domain method for solution of maxwell’s integral equations,

    M. Hano, “Generalized time-domain method for solution of maxwell’s integral equations,” AIP Conference Proceedings, vol. 391, no. 1, pp. 197– 202, 1997

    work page 1997

  30. [30]

    A high-accuracy realization of the Yee algorithm using non-standard finite differences,

    J. Cole, “A high-accuracy realization of the Yee algorithm using non-standard finite differences,” IEEE Transactions on Microwave Theory and Techniques, vol. 45, no. 6, pp. 991–996, 1997

    work page 1997

  31. [31]

    Generalization of the Finite-Difference-Based Time-Domain Methods Using the Method of Moments,

    Z. Chen and S. Luo, “Generalization of the Finite-Difference-Based Time-Domain Methods Using the Method of Moments,” IEEE Trans- actions on Antennas and Propagation , vol. 54, no. 9, pp. 2515–2524, 2006

    work page 2006

  32. [32]

    Time-Domain Finite-Difference and Finite-Element Methods for Maxwell Equations in Complex Media,

    F. L. Teixeira and S. Member, “Time-Domain Finite-Difference and Finite-Element Methods for Maxwell Equations in Complex Media,” IEEE Transactions on Antennas and Propaga- tion, vol. 56, no. 8, pp. 2150–2166, 2008. 13

    work page 2008

  33. [33]

    Geometric Computational Electrodynam- ics with Variational Integrators and Discrete Dif- ferential Forms,

    A. Stern, Y. Tong, M. Desbrun, and J. E. Mars- den, “Geometric Computational Electrodynam- ics with Variational Integrators and Discrete Dif- ferential Forms,” in Geometry, Mechanics, and Dynamics: The Legacy of Jerry Marsden , ser. Fields Institute Communications, D. E. Chang, D. D. Holm, G. Patrick, and T. Ratiu, Eds. New York, NY: Springer New York, 20...

    work page 2015

  34. [34]

    The Hamil- tonian Structure of the Maxwell-Vlasov Equa- tions,

    J. E. Marsden and A. Weinstein, “The Hamil- tonian Structure of the Maxwell-Vlasov Equa- tions,” Physica D, vol. 4, no. 3, p. 394, 1982

    work page 1982

  35. [35]

    On the Construction and Compar- ison of Difference Schemes,

    G. Strang, “On the Construction and Compar- ison of Difference Schemes,” SIAM Journal on Numerical Analysis, vol. 5, no. 3, pp. 506–517, 1968

    work page 1968

  36. [36]

    Periodic Table of the Finite Ele- ments

    D. N. Arnold, “Periodic Table of the Finite Ele- ments.” [Online]. Available: http://www-users. math.umn.edu/∼arnold/femtable/index.html

    work page

  37. [37]

    Whitney forms and their extensions,

    J. Lohi and L. Kettunen, “Whitney forms and their extensions,” Journal of Computational and Applied Mathematics, vol. 393, p. 113520, 2021

    work page 2021

  38. [38]

    On the Product of Semi-Groups of Operators,

    H. F. Trotter, “On the Product of Semi-Groups of Operators,” Proceedings of the American Mathematical Society, vol. 10, no. 4, pp. 545– 551, 1959

    work page 1959

  39. [39]

    Approximate Sparsity Patterns for the Inverse of a Matrix and Preconditioning,

    T. Huckle, “Approximate Sparsity Patterns for the Inverse of a Matrix and Preconditioning,” Technische Universitat Munchen, Tech. Rep. TUM-I9829, 1998

    work page 1998

  40. [40]

    L. N. Trefethen and D. Bau, Numerical linear algebra. Philadelphia: Society for Industrial and Applied Mathematics, 1997

    work page 1997

  41. [41]

    A constrained conjugate gradient method and the solution of linear equations,

    M. Shariff, “A constrained conjugate gradient method and the solution of linear equations,” Computers & Mathematics with Applications , vol. 30, no. 11, pp. 25–37, Dec. 1995. 14

    work page 1995