论文标题
通过量子退火来优化光子晶体表面发射激光器
Towards optimization of photonic-crystal surface-emitting lasers via quantum annealing
论文作者
论文摘要
光子晶体发射激光器(PCSEL)在光子晶体内使用二维(2D)光学共振来进行激光,具有各种出色的功能,例如单模高功率操作和梁极化的任意控制。尽管PCSEL的大多数先前设计都采用了空间均匀的光子晶体,但预计如果有可能优化光子晶体的空间分布,则可以进一步提高激光性能。在本文中,我们通过线性极化的量子退火研究了PCSEL的结构优化。 PCSEL的优化是通过以下三个步骤的迭代进行的:(1)时间依赖性的3D耦合波性能分析激光性能,(2)通过分解机对激光性能进行表述,以及(3)通过量子退火选择最佳解决方案。通过使用这种方法,我们成功地发现了带有带边缘频率和注入电流的非均匀空间分布的高级PCSEL,它同时启用了更高的输出功率,较窄的发散角度和比传统均匀PC的线性偏振比更高的线性极化比。我们的结果可能表明,量子退火的普遍适用性,该量子退火的普遍性主要用于迄今为止特定类型的离散优化问题,用于智能制造领域的各种物理和工程问题。
Photonic-crystal surface-emitting lasers (PCSELs), which utilize a two-dimensional (2D) optical resonance inside a photonic crystal for lasing, feature various outstanding functionalities such as single-mode high-power operation and arbitrary control of beam polarizations. Although most of the previous designs of PCSELs employ spatially uniform photonic crystals, it is expected that lasing performance can be further improved if it becomes possible to optimize the spatial distribution of photonic crystals. In this paper, we investigate the structural optimization of PCSELs via quantum annealing towards high-power, narrow-beam-divergence operation with linear polarization. The optimization of PCSELs is performed by the iteration of the following three steps: (1) time-dependent 3D coupled-wave analysis of lasing performance, (2) formulation of the lasing performance via a factorization machine, and (3) selection of optimal solution(s) via quantum annealing. By using this approach, we successfully discover an advanced PCSEL with a non-uniform spatial distribution of the band-edge frequency and injection current, which simultaneously enables higher output power, a narrower divergence angle, and a higher linear polarization ratio than conventional uniform PCSELs. Our results potentially indicate the universal applicability of quantum annealing, which has been mainly applied to specific types of discrete optimization problems so far, for various physics and engineering problems in the field of smart manufacturing.