EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

MxDiffusion: A Physics-Aware Maxwell’s Law-Guided Diffusion Model Strategy for Inverse Photonic Metasurface Design

Sujoy Mondal iD, Taehyuk Park, Sudipta Biswas, Alan X Wang iD, Wenshan Cai iD

Nano Letters · 15.04.2026 · 26 (14) · 4897-4905

DOI: 10.1021/acs.nanolett.6c00943 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Nano Letters, online erschienen am 06.04.2026. Verfasst von Sujoy Mondal und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 64 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Aeronautics and Space Administration, Division of Materials Research.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We introduce MxDiffusion, a hybrid physics- and data-driven diffusion-based framework that enables the efficient and highly accurate generation of photonic structures from target optical properties. The improved accuracy is achieved through a two-stage generation strategy, in which the first diffusion model is explicitly trained with Maxwell's equation-based loss to embed physical insight directly into the inverse design process, while the second model maps the physically consistent intermediate representation to the final structural geometry with significantly higher fidelity than solely data-driven approaches. The performance of MxDiffusion is validated on two representative applications: gold pattern optimization for random spectral responses and a tunable bandpass filter design based on a phase change material. In both cases, the proposed framework consistently outperforms a conventional data-driven diffusion model benchmark, particularly for out-of-training distribution design targets and highly constrained resonance conditions. These results demonstrate the efficacy and superiority of MxDiffusion as a general physics-guided inverse design paradigm.

Schlagwörter

Stichwörter der Autor:innen

Diffusion ModelInverse DesignMetamaterialsNanophotonicsPhysics-Aware

Autor:innen und Institutionen

  • School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States.
  • Department of Electrical and Computer Engineering, Baylor University, Waco, Texas 76798, United States.
  • School of Materials Science and Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, United States.

ORCID-Profile: Sujoy Mondal, Alan X Wang, Wenshan Cai

Förderung

  • National Aeronautics and Space Administration
  • Division of Materials Research

Zitieren

APA

Mondal, S., Park, T., Biswas, S., Wang, A. X., & Cai, W. (2026). MxDiffusion: A Physics-Aware Maxwell’s Law-Guided Diffusion Model Strategy for Inverse Photonic Metasurface Design. Nano Letters, 26(14), 4897-4905. https://doi.org/10.1021/acs.nanolett.6c00943

Vancouver

Mondal S, Park T, Biswas S, Wang AX, Cai W. MxDiffusion: A Physics-Aware Maxwell’s Law-Guided Diffusion Model Strategy for Inverse Photonic Metasurface Design. Nano Letters. 2026;26(14):4897-4905. doi:10.1021/acs.nanolett.6c00943

BibTeX
@article{Mondal2026,
  author = {Mondal, Sujoy and Park, Taehyuk and Biswas, Sudipta and Wang, Alan X and Cai, Wenshan},
  title = {MxDiffusion: A Physics-Aware Maxwell’s Law-Guided Diffusion Model Strategy for Inverse Photonic Metasurface Design},
  journal = {Nano Letters},
  year = {2026},
  volume = {26},
  number = {14},
  pages = {4897--4905},
  doi = {10.1021/acs.nanolett.6c00943}
}
Kostenlos starten →