AI 中文总结
该研究提出两阶段逆向设计程序，旨在生成二维无序隐身超均匀三价光子网络，实现 TE 和 TM 带隙近最优重叠及小缺陷态密度，获得大 PBGs 并具均匀性，为光子技术操纵电磁波提供新途径。
AI 中文摘要
我们提出了一种两阶段逆向设计程序，用于在二维中生成无序的隐身超均匀三价光子网络，该网络具有各向同性完全光子带隙（PBGs），能在宽频率范围内阻挡任意方向或偏振（TE 或 TM）的光。大多数普通无序系统在系统尺寸增加时无法维持完全 PBGs。唯一已知的例外是通过将隐身超均匀点图案映射到三价网络生成的，但所得网络并非真正的隐身超均匀两相介质。相比之下，我们的两阶段逆向设计旨在使最终网络本身成为隐身超均匀的，实现 TE 和 TM 带隙之间前所未有的近最优重叠以及带边处小的缺陷态密度。我们不仅获得了具有大 PBGs 的单个实现，还在大量集合中实现了显著的均匀性，有效探测了具有 100,000 个顶点的网络。基于集合的带隙宽度与具有相同网络参数的各向异性蜂窝光子晶体的完全 PBG 相当，比之前已知最宽的各向同性完全 PBG 宽近一个数量级。我们的设计可通过增材制造制造，为光子技术操纵电磁波提供了新途径。
英文摘要
We present a two-stage inverse design procedure for producing disordered stealthy hyperuniform trivalent photonic networks in two dimensions with isotropic complete photonic band gaps (PBGs) blocking light regardless of direction or polarization (TE or TM) over a wide frequency range. Most ordinary disordered systems fail to maintain complete PBGs as system size increases. The only known exceptions that remain open in the largest simulations have been generated by mapping stealthy hyperuniform point patterns into trivalent networks. However, the resulting networks are not truly stealthy hyperuniform two-phase media. Although their PBGs remain open, they are relatively narrow due to limited overlap between the TE and TM band gaps and broad band tails caused by localized defect states. By contrast, our two-stage inverse design aims to make the final network itself stealthy hyperuniform, achieving unprecedented near-optimal overlap between the TE and TM band gaps and a small defect state density at the band edges. We obtain not only single realizations with large PBGs, but a striking homogeneity across a large ensemble, effectively probing a network with 100,000 vertices. This ensemble-based band gap is comparable in width to the complete PBG of an anisotropic honeycomb photonic crystal with the same network parameters and nearly an order of magnitude wider than the previously widest known isotropic complete PBGs. Our designs can be fabricated using additive manufacturing, offering new pathways to manipulate electromagnetic waves for photonic technologies.
Comments18 pages, 4 figures
