pith. sign in

arxiv: 2606.25541 · v1 · pith:FOVQUFDPnew · submitted 2026-06-24 · ⚛️ physics.optics · physics.comp-ph

Pseudo-spectral frequency-domain method with background field decomposition and Green's function preconditioner for electromagnetic scattering problem in EUV lithography

Seungjin Lee , Werner Gillijns , Doyun Kim This is my paper

Pith reviewed 2026-06-25 20:18 UTC · model grok-4.3

classification ⚛️ physics.optics physics.comp-ph
keywords electromagnetic scatteringEUV lithographypseudo-spectral frequency-domainGreen's function preconditionerlayered mediacomputational accelerationmask simulation
0
0 comments X

The pith

A reformulation of EUV scattering into a homogeneous background problem with Green's function preconditioning accelerates the pseudo-spectral frequency-domain solver.

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

The paper develops an accelerated method for computing electromagnetic scattering from structures in planarly layered media, as encountered in extreme ultraviolet lithography. It reformulates the problem by decomposing the field into a background component from the layered stack and a scattered component solved on a homogeneous medium. This is combined with a pseudo-spectral frequency-domain discretization and an iterative solver preconditioned by the free-space Green's function. The approach is tested on EUV mask geometries and multilayer mirrors, showing faster convergence than the standard method. A sympathetic reader would care because accurate and fast simulation of EUV masks is critical for semiconductor manufacturing at small scales.

Core claim

The proposed framework reformulates the EUV scattering problem into a scattering problem on a homogeneous background, where the layered media contribution is captured by a recursively updated reflection of the layered stack, solved using the pseudo-spectral frequency-domain method with a free-space Green's function preconditioner, resulting in significant speedup over the conventional approach.

What carries the argument

Background field decomposition paired with a free-space Green's function preconditioner, which allows the iterative solver to converge faster by handling the homogeneous background scattering.

If this is right

  • The method enables faster simulation of complex EUV mask structures.
  • It can be applied to multilayer mirror stacks with improved efficiency.
  • Iterative convergence is expedited without altering the underlying physics model.
  • The framework maintains accuracy while reducing computational cost.

Where Pith is reading between the lines

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

  • This could extend to other layered media problems in optics beyond lithography.
  • Potential for integration with other frequency-domain solvers.
  • Might allow real-time or larger-scale simulations in manufacturing design.

Load-bearing premise

The reformulation into a scattering problem on a homogeneous background with recursively updated reflection accurately captures the electromagnetic contribution of the planarly layered media without significant errors.

What would settle it

A direct comparison of the computed fields or reflection coefficients against a reference solution from a different high-accuracy method on a simple multilayer stack with known analytical solution would show if the speedup comes at the cost of accuracy.

Figures

Figures reproduced from arXiv: 2606.25541 by Doyun Kim, Seungjin Lee, Werner Gillijns.

Figure 1
Figure 1. Figure 1: Schematic representation of the EUV scattering problem. The scattering potential is de [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Decomposition of the EUV scattering problem via the background field framework. (a) [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Numerical convergence analysis for the ∇×∇× differential operator. (a) Isotropic decay￾ing wavefield profile 𝑢 spanning a 30𝜆 isotropic domain. (b), (c) Convergence rate comparisons showing the exponential accuracy of the proposed PSFD scheme versus the algebraic scaling of the conventional finite-difference method with respect to the grid points 𝑁 per axis. The spectral representation of the differential … view at source ↗
Figure 5
Figure 5. Figure 5: 0 1 2 3 4 5 6 𝑥 [𝜆] 0.0 2.5 5.0 7.5 10.0 12.5 15.0 17.5 𝑧 [𝜆] 0.86 0.88 0.90 0.92 0.94 0.96 0.98 1.00 (a) Permittivity distribution (y-cut) 0 1 2 3 4 5 6 𝑥 [𝜆] 0 1 2 3 4 5 6 𝑦 [𝜆] 0.86 0.88 0.90 0.92 0.94 0.96 0.98 1.00 (b) Permittivity distribution (z-cut) [PITH_FULL_IMAGE:figures/full_fig_p007_5.png] view at source ↗
Figure 4
Figure 4. Figure 4: Cross-sectional permittivity maps of the test structure used for verifying the PSFD solver [PITH_FULL_IMAGE:figures/full_fig_p007_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Intensity distributions|𝐸𝑦 | 2 of the scattered electric fields on the 𝑦 = 0 plane, computed by (a) PSFD and (b) the FDFD together with (c) their absolute difference. The white dashed lines indicate the interfaces of the Ta absorber and the layered stack. 0 500 1000 1500 2000 2500 3000 GMRES iteration 10−8 10−7 10−6 10−5 10−4 10−3 10−2 10−1 100 Residual Spectral (1793 it) FDFD (3155 it) tol=1e-08 (a) 0.0 0… view at source ↗
Figure 6
Figure 6. Figure 6: The convergence of the PSFD and FDFD methods for the test problem in Figure 4. The [PITH_FULL_IMAGE:figures/full_fig_p008_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Intensity distributions|𝐸𝑦 | 2 of the scattered electric fields on the 𝑦 = 0 plane, computed by (a) the full-domain PSFD and (b) the background field method together with (c) their absolute difference. The white dashed lines indicate the interfaces of the Ta absorber. As illustrated in [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Comparison between the full-domain PSFD method and the proposed background field [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: The sample mask structure taken from the LithoBench [38] used for the demonstration [PITH_FULL_IMAGE:figures/full_fig_p012_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: The total intensity distribution |𝐄mask + 𝐄bg| 2 of the scattered field on top of the absorber, computed by (a) the background field method without preconditioning and (b) the Lippmann-Schwinger solver together with (c) their absolute difference. formulation against a standard isotropic spectral damping preconditioner defined by the operator 𝐷(𝑘) = min(1, 𝑘2 bg∕𝑘 2 ), which approximates the spectral atten… view at source ↗
Figure 11
Figure 11. Figure 11: The convergence behavior of the iterative solver with different preconditioners for [PITH_FULL_IMAGE:figures/full_fig_p014_11.png] view at source ↗
read the original abstract

We provide an accelerated computational framework to solve electromagnetic scattering problems in planarly layered media arising from extreme ultraviolet (EUV) lithography. To achieve this, we reformulate the EUV scattering problem into a scattering problem on a homogeneous background, in which the electromagnetic contribution of the layered media is captured by a recursively updated reflection of the layered stack. The system is numerically solved by employing the pseudo-spectral frequency-domain method paired with an iterative solver, whose iterative convergence is expedited by a free-space Green's function preconditioner. The proposed framework is evaluated on EUV mask geometries and multilayer mirror stacks, demonstrating a significant speedup over the conventional pseudo-spectral frequency-domain method.

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 / 0 minor

Summary. The manuscript proposes an accelerated framework for electromagnetic scattering in planarly layered media for EUV lithography. It reformulates the problem as scattering on a homogeneous background whose layered contribution is replaced by a recursively updated reflection operator, then solves the resulting system with the pseudo-spectral frequency-domain method, an iterative solver, and a free-space Green's function preconditioner. The approach is evaluated on EUV mask geometries and multilayer mirror stacks and is claimed to deliver significant speedup relative to the conventional pseudo-spectral frequency-domain method.

Significance. If the recursive reflection update preserves accuracy to within the discretization error of the pseudo-spectral scheme and the reported speedup is confirmed with quantitative metrics, the method could reduce the computational burden of EUV mask and multilayer simulations, which are central to semiconductor process development. The background decomposition plus free-space preconditioner is a standard strategy, but its concrete realization here targets a practically important geometry class.

major comments (2)
  1. [Abstract] Abstract: the central claim of 'significant speedup' is asserted without any quantitative metrics, error norms, iteration counts, wall-clock timings, or convergence plots, leaving the primary performance assertion unsupported by evidence.
  2. [Abstract (and implied § on background decomposition)] The reformulation replaces the layered-media contribution by a recursively updated reflection operator applied inside the iterative loop while the preconditioner remains strictly free-space. No a-priori error bound is supplied showing that truncation or discretization mismatch in the recursive coefficients remains below the pseudo-spectral discretization error on the EUV mask geometries; this directly affects the right-hand side and effective operator and is therefore load-bearing for the accuracy-plus-speedup claim.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback on our manuscript. We address each major comment below and will incorporate revisions to strengthen the presentation of results and accuracy analysis.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the central claim of 'significant speedup' is asserted without any quantitative metrics, error norms, iteration counts, wall-clock timings, or convergence plots, leaving the primary performance assertion unsupported by evidence.

    Authors: We agree that the abstract would be strengthened by including quantitative metrics. The full manuscript reports wall-clock timings, iteration counts, error norms, and speedup factors for EUV mask geometries and multilayer stacks in the numerical results section. We will revise the abstract to summarize key metrics (e.g., observed speedup and typical iteration reduction) while keeping it concise. revision: yes

  2. Referee: [Abstract (and implied § on background decomposition)] The reformulation replaces the layered-media contribution by a recursively updated reflection operator applied inside the iterative loop while the preconditioner remains strictly free-space. No a-priori error bound is supplied showing that truncation or discretization mismatch in the recursive coefficients remains below the pseudo-spectral discretization error on the EUV mask geometries; this directly affects the right-hand side and effective operator and is therefore load-bearing for the accuracy-plus-speedup claim.

    Authors: The recursive reflection operator is constructed from the exact Fresnel coefficients of the layered stack and applied without truncation in the continuous formulation; any discretization effects arise only from the pseudo-spectral grid. While the current manuscript relies on numerical verification rather than a formal a-priori bound, comparisons against reference solutions on the tested EUV geometries show that the total error stays within the expected discretization tolerance. We will add a dedicated paragraph in the methods section with supporting error analysis and additional convergence data to explicitly confirm that the recursive operator contribution remains below the discretization error. revision: yes

Circularity Check

0 steps flagged

No circularity: numerical reformulation and preconditioning evaluated on external test cases

full rationale

The paper describes a reformulation of the EUV scattering problem into one on a homogeneous background with a recursive reflection operator for layered media, solved via pseudo-spectral frequency-domain discretization plus free-space Green's preconditioner. Speedup is claimed from direct numerical evaluation on EUV mask and multilayer geometries. No load-bearing step reduces by construction to its own inputs, no fitted parameter is relabeled as a prediction, and no self-citation chain is invoked to justify uniqueness or an ansatz. The derivation chain is self-contained as a standard numerical technique with external verification.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central reformulation rests on the assumption that layered media effects can be exactly captured by recursive reflection in a homogeneous background; no free parameters or invented entities are identifiable from the abstract.

axioms (1)
  • domain assumption The electromagnetic contribution of the planarly layered media can be captured by a recursively updated reflection of the layered stack in the homogeneous background reformulation.
    This is the key step in reformulating the scattering problem as stated in the abstract.

pith-pipeline@v0.9.1-grok · 5649 in / 1216 out tokens · 33174 ms · 2026-06-25T20:18:45.501334+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

47 extracted references · 1 linked inside Pith

  1. [1]

    A.K.-K.Wong,ResolutionEnhancementTechniquesinOpticalLithography,SPIE(2001)

    2001

  2. [2]

    Mack,Fundamental Principles of Optical Lithography: The Science of Microfabrication, JohnWiley&Sons(2007)

    C. Mack,Fundamental Principles of Optical Lithography: The Science of Microfabrication, JohnWiley&Sons(2007)

    2007

  3. [3]

    X.MaandG.R.Arce,ComputationalLithography,JohnWiley&Sons(2011)

    2011

  4. [4]

    Erdmann, D

    A. Erdmann, D. Xu, P. Evanschitzky, V. Philipsen, V. Luong and E. Hendrickx,Characteri- zationandmitigationof3Dmaskeffectsinextremeultravioletlithography,Adv.Opt.Technol. 6(2017)187

    2017

  5. [5]

    A.Erdmann,P.Evanschitzky,G.Bottiglieri,E.vanSettenandT.Fliervoet,3Dmaskeffectsin highNAEUVimaging,Proc.SPIE10957,ExtremeUltraviolet(EUV)LithographyX,109570Z (2019)

    2019

  6. [6]

    H.J.Levinson,ExtremeUltravioletLithography,SPIE(2020)

    2020

  7. [7]

    Yee,Numerical solution of initial boundary value problems involving maxwell’s equations inisotropicmedia,IEEETrans.AntennasPropagat.14(1966)302

    K. Yee,Numerical solution of initial boundary value problems involving maxwell’s equations inisotropicmedia,IEEETrans.AntennasPropagat.14(1966)302

    1966

  8. [8]

    A.Taflove,AdvancesinComputationalElectrodynamics: TheFinite-differenceTime-domain Method,ArtechHouse(1998). 15

    1998

  9. [9]

    Moharam and T.K

    M.G. Moharam and T.K. Gaylord,Rigorous coupled-wave analysis of planar-grating diffrac- tion,J.Opt.Soc.Am.,JOSA71(1981)811

    1981

  10. [10]

    Moharam, D.A

    M.G. Moharam, D.A. Pommet, E.B. Grann and T.K. Gaylord,Stable implementation of the rigorouscoupled-waveanalysisforsurface-reliefgratings: Enhancedtransmittancematrixap- proach,J.Opt.Soc.Am.A,JOSAA12(1995)1077

    1995

  11. [11]

    Q.H.Liu,ThePSTDalgorithm: Atime-domainmethodrequiringonlytwocellsperwavelength, Microw.Opt.Technol.Lett.15(1997)158

    1997

  12. [12]

    L.F.Abbott,IntroductiontotheBackgroundFieldMethod,ActaPhys.Polon.B13(1982)33

    1982

  13. [13]

    W.C.Chew,WavesandFieldsinInhomogeneousMedia,Wiley-IEEEPress(1995)

    1995

  14. [14]

    W.Chew,E.Michielssen,J.M.SongandJ.M.Jin,FastandEfficientAlgorithmsinComputa- tionalElectromagnetics,ArtechHouse(2001)

    2001

  15. [15]

    Colton and R

    D. Colton and R. Kress,Inverse Acoustic and Electromagnetic Scattering Theory, Springer (2020)

    2020

  16. [16]

    M.BornandE.Wolf,PrinciplesofOptics: ElectromagneticTheoryofPropagation,Interference andDiffractionofLight,7thed.,CambridgeUniversityPress(2013)

    2013

  17. [17]

    J.D.Jackson,ClassicalElectrodynamics,3rded.,Wiley(1998)

    1998

  18. [18]

    Chew,Response of a source on top of a vertically stratified half-space,IEEE Transactions onAntennasandPropagation33(1985)649

    W. Chew,Response of a source on top of a vertically stratified half-space,IEEE Transactions onAntennasandPropagation33(1985)649

    1985

  19. [19]

    T.G.MackayandA.Lakhtakia,TheTransfer-MatrixMethodinElectromagneticsandOptics, Springer(2020)

    2020

  20. [20]

    T.V.Pistor,K.AdamandA.Neureuther,Rigoroussimulationofmaskcornereffectsinextreme ultravioletlithography,J.Vac.Sci.Technol.B16(1998)3449

    1998

  21. [21]

    J.P.Boyd,ChebyshevandFourierSpectralMethods,2nded.,DoverPublications(2001)

    2001

  22. [22]

    Fornberg,A Practical Guide to Pseudospectral Methods, Cambridge University Press (1998)

    B. Fornberg,A Practical Guide to Pseudospectral Methods, Cambridge University Press (1998)

    1998

  23. [23]

    Liu,Large-scale simulations of electromagnetic and acoustic measurements using the pseudospectraltime-domain(PSTD)algorithm,IEEETransactionsonGeoscienceandRemote Sensing37(1999)917

    Q.H. Liu,Large-scale simulations of electromagnetic and acoustic measurements using the pseudospectraltime-domain(PSTD)algorithm,IEEETransactionsonGeoscienceandRemote Sensing37(1999)917

    1999

  24. [24]

    D.Song,L.YuanandY.Y.Lu,Fourier-matchingpseudospectralmodalmethodfordiffraction gratings,J.Opt.Soc.Am.A,JOSAA28(2011)613

    2011

  25. [25]

    Munro, D

    P.R.T. Munro, D. Engelke and D.D. Sampson,A compact source condition for modelling fo- cusedfieldsusingthepseudospectraltime-domainmethod,Opt.Express,OE22(2014)5599

    2014

  26. [26]

    Berenger,Aperfectlymatchedlayerfortheabsorptionofelectromagneticwaves,Journal ofComputationalPhysics114(1994)185

    J.-P. Berenger,Aperfectlymatchedlayerfortheabsorptionofelectromagneticwaves,Journal ofComputationalPhysics114(1994)185

    1994

  27. [27]

    I.A.Makhotkin,M.Wu,V.Soltwisch,F.ScholzeandV.Philipsen,Refinedextremeultraviolet maskstackmodel,J.Opt.Soc.Am.A,JOSAA38(2021)498. 16

    2021

  28. [28]

    Saad and M.H

    Y. Saad and M.H. Schultz,GMRES: A Generalized Minimal Residual Algorithm for Solving NonsymmetricLinearSystems,SIAMJ.Sci.andStat.Comput.7(1986)856

    1986

  29. [29]

    Y.Saad,IterativeMethodsforSparseLinearSystems,2nded.,SIAM(2003)

    2003

  30. [30]

    Barrett, M

    R. Barrett, M. Berry, T.F. Chan, J. Demmel, J. Donato, J. Dongarra et al.,Templates for the SolutionofLinearSystems: BuildingBlocksforIterativeMethods,SIAM(1994)

    1994

  31. [31]

    O.Nevanlinna,ConvergenceofIterationsforLinearEquations,Birkhäuser,Basel(1993)

    1993

  32. [32]

    A.Greenbaum,IterativeMethodsforSolvingLinearSystems,SIAM(1997)

    1997

  33. [33]

    J.LiesenandZ.Strakos,KrylovSubspaceMethods: PrinciplesandAnalysis,OxfordUniversity Press(2012)

    2012

  34. [34]

    Vainikko,Fast Solvers of the Lippmann-Schwinger Equation, in R.P

    G. Vainikko,Fast Solvers of the Lippmann-Schwinger Equation, in R.P. Gilbert, J. Kajiwara, Y.S. Xu and R.P. Gilbert (eds.)Direct and Inverse Problems of Mathematical Physics, vol. 5, Springer(2000)

    2000

  35. [35]

    Comput.Phys.323(2016)191[1604.03155]

    F.Vico,L.GreengardandM.Ferrando,Fastconvolutionwithfree-spaceGreen’sfunctions,J. Comput.Phys.323(2016)191[1604.03155]

    Pith/arXiv arXiv 2016

  36. [36]

    T.-a.Pham,E.Soubies,A.Ayoub,J.Lim,D.PsaltisandM.Unser,Three-DimensionalOptical Diffraction Tomography with Lippmann-Schwinger Model,IEEE Transactions on Computa- tionalImaging,6(2020)727

    2020

  37. [37]

    R.Kress,LinearIntegralEquations,2nded.,Springer(1999)

    1999

  38. [38]

    S.Zheng,H.Yang,B.Zhu,B.YuandM.Wong,LithoBench: BenchmarkingAIComputational Lithography for Semiconductor Manufacturing,Advances in Neural Information Processing Systems36(2023)30243

    2023

  39. [39]

    Thakare, A

    D. Thakare, A. Delabie and V. Philipsen,Optimizing EUV imaging metrics as a function of absorberthicknessandilluminationsource: SimulationcasestudyofTa-Coalloy,Proc.SPIE 12472,37thEuropeanMaskandLithographyConference,124720A(2022)

    2022

  40. [40]

    SPIE 5256, 23rd Annual BACUS Symposium on Pho- tomaskTechnology(2003)

    R.Windpassinger,N.Rosenkranz,T.Scherubl,P.Evanschitzky,A.ErdmannandA.Zibold, EUV mask simulation for AIMS,Proc. SPIE 5256, 23rd Annual BACUS Symposium on Pho- tomaskTechnology(2003)

    2003

  41. [41]

    W.CaiandT.Yu,FastCalculationsofDyadicGreen’sFunctionsforElectromagneticScattering inaMultilayeredMedium,J.Comput.Phys.165(2000)1

    2000

  42. [42]

    Cai,ComputationalMethodsforElectromagneticPhenomena: ElectrostaticsinSolvation, Scattering,andElectronTransport,CambridgeUniversityPress(2013)

    W. Cai,ComputationalMethodsforElectromagneticPhenomena: ElectrostaticsinSolvation, Scattering,andElectronTransport,CambridgeUniversityPress(2013)

    2013

  43. [43]

    H. Yuan, B. Wang, W. Zhang and W. Cai,Fast Multipole Method for Maxwell’s Equations in LayeredMedia,arXiv:2507.18491[math.NA](2025)

    arXiv 2025

  44. [44]

    60(1985)187

    V.Rokhlin,Rapidsolutionofintegralequationsofclassicalpotentialtheory,J.Comput.Phys. 60(1985)187. 17

    1985

  45. [45]

    Greengard and V

    L. Greengard and V. Rokhlin,A fast algorithm for particle simulations,J. Comput. Phys.73 (1987)325

    1987

  46. [46]

    Carrier, L

    J. Carrier, L. Greengard and V. Rokhlin,A Fast Adaptive Multipole Algorithm for Particle Simulations,SIAMJ.Sci.andStat.Comput.9(1988)669

    1988

  47. [47]

    Ergul and L

    O. Ergul and L. Gurel,TheMultilevelFastMultipoleAlgorithm(MLFMA)forSolvingLarge- ScaleComputationalElectromagneticsProblems,JohnWiley&Sons(2014). 18

    2014