arxiv: 2605.09470 · v1 · submitted 2026-05-10 · ⚛️ physics.optics · physics.comp-ph

Recognition: 2 theorem links

· Lean Theorem

Accuracy assessment of scalar wave propagation methods for diffractive optics design: from thin elements to thick binary grating

Nicolas Barr\'e

Authors on Pith no claims yet

Pith reviewed 2026-05-12 04:58 UTC · model grok-4.3

classification ⚛️ physics.optics physics.comp-ph

keywords diffractive opticsbinary gratingsthin-element approximationbeam propagation methodwave propagation methodaccuracy assessmentFourier modal methodscalar methods

0 comments

The pith

Accuracy maps in frequency-thickness space mark where thin-element, BPM, and WPM methods match rigorous results for binary gratings.

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

The paper compares the thin-element approximation, beam propagation method, and wave propagation method to the rigorous Fourier modal method on randomly generated binary diffractive gratings. It computes the overlap of transmitted fields across ranges of spatial frequency cutoffs and grating thicknesses. The comparisons are condensed into accuracy maps plotted in the spatial frequency-thickness parameter space. These maps identify the regions where each scalar method remains reliable and supply rules of thumb for selecting a forward model inside inverse design pipelines for diffractive optics. A reader cares because the wrong model choice can produce inefficient or inaccurate optical designs.

Core claim

The central claim is that accuracy maps in the spatial frequency-thickness parameter space reveal the domains of validity for the thin-element approximation, the beam propagation method, and the wave propagation method when applied to binary diffractive gratings, using the Fourier modal method as reference, and thereby provide practical guidelines for choosing forward models in inverse design pipelines.

What carries the argument

Accuracy maps in the spatial frequency-thickness parameter space that display transmitted-field overlap between each scalar method and the rigorous Fourier modal method reference.

If this is right

The thin-element approximation remains accurate primarily for thin gratings at low spatial frequencies.
Beam propagation and wave propagation methods maintain accuracy for thicker gratings and higher frequencies than the thin-element approximation.
Design pipelines can use the maps to select the cheapest forward model that still meets a target accuracy.
Random binary gratings form a sufficient test ensemble for mapping method validity.

Where Pith is reading between the lines

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

The maps could be interpolated into a continuous decision surface that an optimizer queries at each iteration.
Non-binary or multilevel gratings might shift the validity boundaries, requiring separate maps.
Experimental confirmation on fabricated samples would test whether field overlap predicts end-to-end system performance.

Load-bearing premise

That overlap of the transmitted field is a sufficient and representative accuracy metric for the needs of diffractive optics inverse design, and that randomly generated binary gratings adequately cover the relevant design space.

What would settle it

Fabricate binary gratings designed under each scalar method inside and outside the claimed validity regions, then measure actual diffraction efficiency or image quality in an optical setup and check whether measured performance tracks the predicted overlap values.

Figures

Figures reproduced from arXiv: 2605.09470 by Nicolas Barr\'e.

**Figure 1.** Figure 1: illustrates representative binary grating profiles for increasing spatial frequency cutoff 𝑓𝑐 at fixed thickness ℎ = 3𝜆. As 𝑓𝑐 increases, the lateral feature size decreases and the grating acquires finer structure, generating larger diffraction angles. At low 𝑓𝑐, the grating consists of smooth, large-scale features that closely satisfy the TEA assumptions. At high 𝑓𝑐, sharp sub-wavelength features dominate… view at source ↗

**Figure 2.** Figure 2: Transmitted intensity (top row) and phase (bottom row) computed by TEA, BPM, WPM, and the FMM reference for a representative binary grating at 𝜃max ≈ 16.5 ◦ and ℎ = 3𝜆. Each intensity panel is normalized to its own maximum. The corresponding field overlaps with the FMM reference are 𝜂TEA = 75.1%, 𝜂BPM = 89.6%, and 𝜂WPM = 97.3%. The accuracy maps in [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Figure 3: Accuracy maps showing the mean field overlap 𝜂 between each scalar method and the FMM reference, as a function of spatial frequency cutoff 𝑓𝑐 (expressed as maximum diffraction angle 𝜃max) and grating thickness ℎ (in units of 𝜆), averaged over 10 random binary grating realizations. The red contour indicates the 𝜂 = 0.90 level. Binary gratings with 𝑛1 = 1.5, 𝑛2 = 1.0, 𝜆 = 532 nm. at lateral index discontinui… view at source ↗

read the original abstract

We present a systematic accuracy assessment of the thin-element approximation (TEA), the beam propagation method (BPM), and the wave propagation method (WPM) for binary diffractive gratings, using the rigorous Fourier modal method (FMM) as a reference. Random binary gratings are generated over a range of spatial frequency cutoffs and thicknesses, and the transmitted field overlap between each scalar method and the reference is measured. The results are summarized as accuracy maps in the spatial frequency-thickness parameter space, revealing the domain of validity of each method and providing practical guidelines for the choice of forward model in diffractive optics inverse design pipelines.

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.

Desk Editor's Note private letter to a colleague

This paper maps where TEA, BPM, and WPM stay close to FMM on random binary gratings via field overlap, which gives usable rules of thumb but leaves the metric and sampling choices open to question for actual inverse-design work.

read the letter

The core result is a set of accuracy maps in the spatial-frequency versus thickness plane that show the regions where the thin-element approximation, beam propagation method, and wave propagation method produce transmitted fields that overlap well with a Fourier modal method reference. The authors generate random binary gratings, compute the overlap for each scalar method, and turn the numbers into practical guidelines for choosing a forward model inside diffractive-optics design loops. That systematic sweep is the new piece; earlier work had compared these methods but not in this compact, map-like form over the relevant parameter space. The comparison itself is straightforward and uses the right reference method, so the maps are at least reproducible in principle and give designers a concrete starting point for when a scalar approximation is safe enough. The soft spot is exactly the one the stress-test flags. Field overlap is a convenient scalar, but inverse-design pipelines usually care about diffraction-order efficiencies, local phase fidelity, or tolerance to fabrication errors; the paper does not show that high overlap guarantees low error on those quantities. Random binary gratings also lack the spatial correlations that appear once you optimize a structure, so the reported validity domains could shift for the gratings that actually matter. If the full text spells out the grating-generation statistics and the precise overlap definition, that helps, but the central claim still rests on an untested proxy. This is the sort of paper that belongs in the reading group for anyone who runs diffractive-optics simulations day to day. It is not foundational, but the output is directly usable and the comparison is honest, so it deserves a serious referee rather than a desk reject.

Referee Report

2 major / 0 minor

Summary. The manuscript claims to provide a systematic accuracy assessment of the thin-element approximation (TEA), beam propagation method (BPM), and wave propagation method (WPM) against the rigorous Fourier modal method (FMM) reference for binary diffractive gratings. Random binary gratings are generated over ranges of spatial frequency cutoffs and thicknesses; transmitted-field overlap is computed for each scalar method; results are presented as accuracy maps in the spatial-frequency–thickness parameter space to delineate validity domains and supply practical guidelines for choosing forward models in diffractive-optics inverse-design pipelines.

Significance. If the maps prove robust, the work would supply concrete, parameter-space guidelines that could help practitioners select computationally efficient scalar propagators without sacrificing essential accuracy in inverse-design loops. The use of an independent rigorous reference (FMM) and a direct numerical comparison strategy is standard and strengthens the assessment.

major comments (2)

Abstract: the central claim that the accuracy maps supply 'practical guidelines for the choice of forward model in diffractive optics inverse design pipelines' rests on two untested assumptions—(i) that transmitted-field overlap is a sufficient proxy for the errors that actually degrade design performance (e.g., diffraction-order efficiencies, localized phase fidelity, or fabrication tolerance), and (ii) that randomly generated binary gratings adequately sample the structures encountered under optimization. Neither correlation nor coverage is demonstrated, rendering the maps' applicability to inverse design only partially supported.
Results (accuracy maps): no quantitative comparison is reported between the overlap metric and order-specific or spatially localized error measures, nor is any test performed on gratings produced by actual inverse-design optimization (which typically exhibit spatial correlations absent from the random ensemble). These omissions are load-bearing for the stated utility of the maps.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the careful and constructive review of our manuscript. The comments correctly identify limitations in how directly our results support claims about inverse-design utility. We address each point below and will revise the manuscript accordingly to clarify scope and limitations.

read point-by-point responses

Referee: Abstract: the central claim that the accuracy maps supply 'practical guidelines for the choice of forward model in diffractive optics inverse design pipelines' rests on two untested assumptions—(i) that transmitted-field overlap is a sufficient proxy for the errors that actually degrade design performance (e.g., diffraction-order efficiencies, localized phase fidelity, or fabrication tolerance), and (ii) that randomly generated binary gratings adequately sample the structures encountered under optimization. Neither correlation nor coverage is demonstrated, rendering the maps' applicability to inverse design only partially supported.

Authors: We agree that the manuscript relies on transmitted-field overlap for random binary gratings without demonstrating explicit correlations to design-specific metrics such as diffraction efficiencies or tests on optimized structures. The overlap metric was selected as it quantifies overall field fidelity, which underpins any forward model used in gradient-based design. Random generation was chosen to enable dense, unbiased sampling of the spatial-frequency–thickness space. We will revise the abstract to specify that the guidelines are derived from field-overlap accuracy on random ensembles and add a paragraph in the Discussion acknowledging that further validation against optimization-specific errors and correlated structures would strengthen applicability. revision: yes
Referee: Results (accuracy maps): no quantitative comparison is reported between the overlap metric and order-specific or spatially localized error measures, nor is any test performed on gratings produced by actual inverse-design optimization (which typically exhibit spatial correlations absent from the random ensemble). These omissions are load-bearing for the stated utility of the maps.

Authors: The current work does not include quantitative comparisons of overlap to order efficiencies or localized errors, nor tests on inverse-design-optimized gratings. Such additions would require a separate, computationally intensive study focused on optimization loops. Our emphasis is on systematic parameter-space coverage with random gratings to reveal general validity domains. In revision we will insert a brief discussion relating overlap deviations to expected impacts on diffraction orders and noting that random ensembles lack the spatial correlations typical of optimized designs, thereby qualifying the maps' direct use in inverse-design pipelines. revision: partial

Circularity Check

0 steps flagged

No circularity: direct numerical comparison against independent reference

full rationale

The paper performs an empirical accuracy assessment by generating random binary gratings across spatial-frequency and thickness ranges, computing transmitted-field overlap between scalar methods (TEA, BPM, WPM) and the rigorous FMM reference, and summarizing results as validity maps. No derivations, parameter fits, or self-citations are load-bearing; the maps are direct simulation outputs. The assessment is self-contained against external benchmarks with no reduction of claims to inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

The work rests on standard domain assumptions in computational optics rather than new free parameters or invented entities.

axioms (2)

domain assumption The Fourier modal method supplies the exact reference solution for the chosen grating geometries
Invoked when overlap is measured against FMM.
domain assumption Field overlap is an appropriate scalar metric for judging suitability in inverse-design pipelines
Central to how accuracy is quantified and mapped.

pith-pipeline@v0.9.0 · 5398 in / 1192 out tokens · 51506 ms · 2026-05-12T04:58:57.414996+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

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

IndisputableMonolith/Foundation/RealityFromDistinction.lean reality_from_one_distinction unclear
accuracy maps in the spatial frequency-thickness parameter space... practical guidelines for the choice of forward model in diffractive optics inverse design pipelines
IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear
transmitted field overlap η between each scalar method and the FMM reference

Reference graph

Works this paper leans on

8 extracted references · 8 canonical work pages

[1]

Journal of Open Source Software , keywords =

Barré, Nicolas , title =. Journal of Open Source Software , keywords =. doi:10.21105/joss.09734 , year =

work page doi:10.21105/joss.09734
[2]

Differentiable wave propagation method for shape optimization of freeform optics beyond the paraxial approximation , volume =

Nicolas Barr\'. Differentiable wave propagation method for shape optimization of freeform optics beyond the paraxial approximation , volume =. Optics Letters , number =. 2025 , doi =

work page 2025
[3]

2023 , eprint=

Fourier modal method for inverse design of metasurface-enhanced micro-LEDs , author=. 2023 , eprint=

work page 2023
[4]

Schmidt and T

S. Schmidt and T. Tiess and S. Schr\". Wave-optical modeling beyond the thin-element approximation , volume =. Optics Express , number =. 2016 , doi =

work page 2016
[5]

Inverse design and demonstration of high-performance wide-angle diffractive optical elements , volume =

Dong Cheon Kim and Andreas Hermerschmidt and Pavel Dyachenko and Toralf Scharf , journal =. Inverse design and demonstration of high-performance wide-angle diffractive optical elements , volume =. 2020 , doi =

work page 2020
[6]

Computer Physics Communications , author=

S4 : A free electromagnetic solver for layered periodic structures , journal =. 2012 , issn =. doi:10.1016/j.cpc.2012.04.026 , author =

work page doi:10.1016/j.cpc.2012.04.026 2012
[7]

2010 , issn =

Meep: A flexible free-software package for electromagnetic simulations by the FDTD method , journal =. 2010 , issn =. doi:10.1016/j.cpc.2009.11.008 , author =

work page doi:10.1016/j.cpc.2009.11.008 2010
[8]

New method for nonparaxial beam propagation , volume =

Anurag Sharma and Arti Agrawal , journal =. New method for nonparaxial beam propagation , volume =. 2004 , doi =

work page 2004