Polarizing beam splitter is designed with a broadband and wide range of incident angle. Polarizing beam splitter designed here is a kind of double periodic subwavelength medium-metal grating, which consisted of silicon dioxide as the substrate, magnesium fluoride (MgF2) as medium material and silver for the grid lines. It has the polarization splitting function of TE reflection and TM transmission. Based on the rigorous coupled-wave analysis(RCWA) and the continuous optimization of the structure parameters, the polarization beam splitter has high polarization conversion efficiency, high extinction ratio and a wide tolerance of incident angle in the near infrared band (1μm − 3μm) .The simulation results show that the reflection efficiency of TE polarized light and the transmission efficiency of TM polarized light are both higher than 96%, and the reflection extinction ratio and transmission extinction ratio are greater than 17dB and 28dB respectively. When the incident angle of incident light is from -80° to 80°, the reflection efficiency of TE polarized light is over 96%; when the incident angle is from -40°to 40°, the transmission efficiency of TM polarized light is over 90%. The reflection extinction ratio exceeds 17dB, and the transmission extinction ratio exceeds 35dB in the incident wavelength of 1550nm.The designed polarizing beam splitter is expected to be used in optical communication, optical storage, optical sensing and other fields for light modulation and control.
Inverse design is one of the most important design methods of nanophotonic devices. In recent years, with the rapid development of deep learning technique and applications, deep learning assisted inverse design method has been introduced into the field of nanophotonic device design. In this work, by combining the direct binary search method with multilayer convolutional neural networks, we present the inverse design of a wavelength demultiplexer which has 1352 design variables. The dropout strategy has been employed to avoid overfitting in training the inverse design model. The simulation results indicate that the trained CNN can both efficiently forward predict the spectrum and inverse design the structure.
Using the principle of Surface Plasmons Polaritons, we propose a graphene broadband terahertz absorption structure that achieves the tunable absorption of electromagnetic waves. In our absorption structure, the method of patterning graphene is used to realize continuous broadband absorption from 0.5THz to 2.1THz. The absorption which is more than 50% reaches 1.1THZ, especially the structure designed here has three plasmonic resonance peaks which above 98% at 0.79THz, 1.18THz and 1.35THz, respectively. In addition, the symmetry in the pattern design consider that our absorption structure is not sensitive to the polarization and incident angle. Due to a series of excellent characteristics of the absorption structure, it may play an important role in the field of aircraft stealth, absorber, and light wave modulation.