Yongmin Liu1
Northeastern University1
Over the past decades, we have witnessed tremendous progress and success of photonic metasurfaces. By tailoring the geometry of the building blocks of metasurfaces and engineering their spatial distribution, we can control the amplitude, polarization state, phase and trajectory of light in an almost arbitrary manner. For practical applications such as high-capacity optical display, information encryption, and data storage, it is crucial to encode distinct functionalities into a single meta-device and increase the information channels. In this talk, I will present our recent works that aim to push the multiplexing limit in photonic metasurfaces through both physics-guided and data-driven approaches. In the first part of my talk, I will show that we can break the fundamental limit of polarization multiplexing capacity of metasurfaces by introducing the engineered noise to the precise solution of Jones matrix elements [<b>Science</b> 379, 294 (2023)]. We experimentally demonstrate up to 11 independent holographic images using a single metasurface illuminated by visible light with different polarizations. To the best of our knowledge, it is the highest capacity reported for polarization multiplexing. Combining our noise engineering method with the position multiplexing scheme, we design and demonstrate another metasurface that can generate 36 distinct images, forming a holographic keyboard pattern. In the second part of my talk, I will show that machine learning enables us to accelerate the development of complex metasurfaces (and other photonic structures) with high efficiency, accuracy and fidelity [<b>Nature Photonics</b> 15, 77 (2021)]. The developed machine learning models, after training, can evaluate the optical responses of metasurfaces and inversely design them in less than seconds. We propose to embed machine learning models in both gradientbased and nongradient optimization loops for the automatic implementation of multifunctional metasurfaces [<b>Advanced Materials</b> 34, 2110022 (2022)]. Fundamentally different from the traditional two-step approach that separates phase retrieval and meta-atom structural design, the proposed end-to-end framework facilitates full exploitation of the prescribed design space and pushes the multifunctional design capacity to its physical limit. With a single-layer structure that can be readily fabricated, metasurface focusing lenses and holograms are experimentally demonstrated in the near-infrared region. They show up to eight controllable responses subjected to different combinations of working frequencies and linear polarization states.