Chenkai Mao

All Publications

• Accurate and scalable deep Maxwell solvers. Proceedings of the National Academy of Sciences of the United States of America Mao, C., Fan, J. A. 2026; 123 (18): e2530330123

  Abstract

  Neural networks have promise as surrogate partial differential equation (PDE) solvers, but it remains a challenge to use these concepts to solve problems with high accuracy and scalability. In this work, we show that neural network surrogates can combine with iterative algorithms to accurately solve PDE problems featuring different scales, resolutions, and boundary conditions. We develop a subdomain neural operator model that supports arbitrary Robin-type boundary condition inputs, and we show that it can be utilized as a flexible preconditioner to iteratively solve subdomain problems with bounded accuracy. We further show that our subdomain models can facilitate the construction of global coarse spaces to enable accelerated, large scale PDE problem solving based on iterative multilevel domain decomposition. With two-dimensional Maxwell's equations as a model system, we train a single network to simulate large scale problems with different sizes, resolutions, wavelengths, and dielectric media distribution. We further demonstrate the utility of our platform in performing the accurate inverse design of multiwavelength nanophotonic devices. Our work presents a promising path to building accurate and scalable multiphysics surrogate solvers for large practical problems.

  View details for DOI 10.1073/pnas.2530330123

  View details for PubMedID 42044333

• Freeform Mode-Engineered Metasurfaces. Nano letters Jiang, Z., Dai, T., Guo, S., Sohag, S. H., Shao, Y., Mao, C., Alù, A., Fan, J. A., Zhou, Y. 2026

  Abstract

  Nanophotonic technologies inherently rely on tailoring light-matter interactions through the excitation and interference of deeply confined optical resonances. However, existing concepts in optical mode engineering remain heuristic and are challenging to extend toward complex and multifunctional resonant phenomena. We introduce an inverse design framework that optimizes near-field distributions, ideally suited to tailoring Mie-type modes within dielectric nanophotonic structures, and we demonstrate its application to the discovery of new classes of nonlocal metasurfaces. We show that freeform nonlocal metasurfaces supporting accidental bound states in the continuum can be readily optimized for tailored illumination conditions, modal properties, and quality factors. We further generalize the framework to higher-order and multifunctional mode engineering and experimentally demonstrate freeform planar nonlocal multiwavelength and chiral metasurfaces. Our versatile framework for freeform mode engineering has applications in broad high-quality-factor nanophotonic platforms relevant to sensing, nonlinear optics, optomechanics, and quantum information processing.

  View details for DOI 10.1021/acs.nanolett.5c06075

  View details for PubMedID 41805361

• 3D nanolithography with metalens arrays and spatially adaptive illumination. Nature Gu, S., Mao, C., Guell Izard, A., Sadana, S., Terrel-Perez, D., Mettry-Yassa, M., Choi, W., Zhou, W., Yan, H., Zhou, Z., Massey, T., Abelson, A., Zhou, Y., Huang, S., Daraio, C., Tumkur, T. U., Fan, J. A., Xia, X. 2025; 648 (8094): 591-599

  Abstract

  The growing demand for advanced materials, miniaturized devices and integrated microsystems calls for the reliable fabrication of complex, multiscale, three-dimensional (3D) architectures, a need increasingly addressed through light-based and laser-based processes. However, owing to the field-of-view (FOV) limitations of conventional imaging optics, existing 3D laser nanofabrication techniques face fundamental challenges in throughput, proximity error and stitching defects on the path to scaling. Here we present a scalable 3D nanofabrication platform that uses a metalens-generated focal spot array to parallelize two-photon lithography (TPL)1 beyond centimetre-scale write field areas. Metalenses are ideally suited for producing submicron-scale focal spots for high-throughput nanolithography, as they uniquely feature large numerical apertures (NAs), immersion media compatibility and large-scale manufacturability. We experimentally demonstrate a printing system that uses a 12-cm2 metalens array to produce more than 120,000 cooperative focal spots, corresponding to a throughput exceeding 108 voxels s-1. By programmatically patterning the focal spot array using a spatial light modulator (SLM), an adaptive parallel printing strategy is developed for precise greyscale linewidth modulation and choreographed printing of semiperiodic and fully aperiodic 3D geometries. We demonstrate parallel printing of replicated microstructures (>50 M microparticles per day), centimetre-scale 3D architectures with feature sizes down to 113 nm, and photonic and mechanical metamaterials. This work demonstrates the potential of 3D nanolithography towards wafer-scale production, showing how TPL could be used at scale for applications in microelectronics2, biomedicine3, quantum technology4 and high-energy laser targets5,6.

  View details for DOI 10.1038/s41586-025-09842-x

  View details for PubMedID 41407898

  View details for PubMedCentralID 11525192

• A multi-agentic framework for real-time, autonomous freeform metasurface design. Science advances Lupoiu, R., Shao, Y., Dai, T., Mao, C., Edée, K., Fan, J. A. 2025; 11 (44): eadx8006

  Abstract

  Innovation in nanophotonics currently relies on human experts who synergize specialized knowledge in photonics and coding with simulation and optimization algorithms, entailing design cycles that are time-consuming, computationally demanding, and frequently suboptimal. We introduce MetaChat, a multi-agentic design framework that can translate semantically described photonic design goals into high-performance, freeform device layouts in an automated, nearly real-time manner. Multistep reasoning is enabled by our Agentic Iterative Monologue paradigm, which coherently interfaces agents with code-based tools, other specialized agents, and human designers. Design acceleration is facilitated by Feature-wise Linear Modulation-conditioned Maxwell surrogate solvers that support the generalized evaluation of metasurface structures. We use freeform dielectric metasurfaces as a model system and demonstrate with MetaChat the design of multiobjective, multiwavelength metasurfaces orders of magnitude faster than conventional methods. These concepts present a scientific computing blueprint for using specialist design agents, surrogate solvers, and human interactions to drive multiphysics innovation and discovery.

  View details for DOI 10.1126/sciadv.adx8006

  View details for PubMedID 41171904

• Shaping freeform nanophotonic devices with geometric neural parameterization NPJ COMPUTATIONAL MATERIALS Dai, T., Shao, Y., Mao, C., Wu, Y., Azzouz, S., Zhou, Y., Fan, J. A. 2025; 11 (1)

  View details for DOI 10.1038/s41524-025-01752-w

  View details for Web of Science ID 001546112400001

• Inverse-designed metasurfaces with facile fabrication parameters JOURNAL OF OPTICS Zhou, Y., Shao, Y., Mao, C., Fan, J. A. 2024; 26 (5)

  View details for DOI 10.1088/2040-8986/ad33a7

  View details for Web of Science ID 001188548100001

• Large-Area, High-Numerical-Aperture, Freeform Metasurfaces LASER & PHOTONICS REVIEWS Zhou, Y., Mao, C., Gershnabel, E., Chen, M., Fan, J. A. 2024

  View details for DOI 10.1002/lpor.202300988

  View details for Web of Science ID 001163678600001

• Towards General Neural Surrogate Solvers with Specialized Neural Accelerators Proceedings of the 41st International Conference on Machine Learning Mao, C., Lupoiu, R., Dai, T., Chen, M., Fan, J. 2024
• Reparameterization Approach to Gradient-Based Inverse Design of Three-Dimensional Nanophotonic Devices ACS PHOTONICS Gershnabel, E., Chen, M., Mao, C., Wang, E. W., Lalanne, P., Fan, J. A. 2022

  View details for DOI 10.1021/acsphotonics.2c01160

  View details for Web of Science ID 000874561200001

• High Speed Simulation and Freeform Optimization of Nanophotonic Devices with Physics-Augmented Deep Learning ACS PHOTONICS Chen, M., Lupoiu, R., Mao, C., Huang, D., Jiang, J., Lalanne, P., Fan, J. A. 2022

  View details for DOI 10.1021/acsphotonics.2c00876

  View details for Web of Science ID 000848111900001

• WaveY-Net: Physics-Augmented Deep Learning for High-Speed Electromagnetic Simulation and Optimization Chen, M., Lupoiu, R., Mao, C., Huang, D., Jiang, J., Lalanne, P., Fan, J. A. edited by Chang-Hasnain, C. J., Fan, J. A., Zhou, W. SPIE-INT SOC OPTICAL ENGINEERING. 2022

  View details for DOI 10.1117/12.2612418

  View details for Web of Science ID 000836330700011