The focal plane polarimeter (DoFP) polarization imaging is one of the most popular polarization imaging methods. It works in real time, has a simple optical path, and can be easily integrated. The polarization filter is an important part of the DoFP. Typical polarization filter fabrication methods include photolithography, focused ion-beam etching, and nanoimprint lithography. A filter prepared using these methods has the advantages of high resolution and favorable imaging effects. However, these methods have significant fabrication difficulties, complex processing steps, and strict requirements for the operating environment. Moreover, the prepared polarizer exhibits a low extinction ratio. Compared to the above fabrication processes, laser processing has the advantages of a simple process and flexible operation and does not require a mask. In this study, a fabrication process for a multi-directional polarization filter is proposed. This process employs a picosecond laser to ablate the unidirectional subwavelength metal grating polarizer to form a regular polarization array and then obtain a multi-directional polarization filter by cementation. This method can achieve large-area fabrication of polarization grating arrays with a high extinction ratio.
Commercial subwavelength metal grating polarizers are ablated by a picosecond laser to remove the polarization effects in the ablation zone and obtain regular three-directional polarization arrays. A polarization filter is obtained by alignment and cementation in their respective directions. During the fabrication process, the laser power is strictly controlled to avoid excessive ablation on the TAC substrate and retain the high transmittance of the substrate. Meanwhile, residue deposition during laser ablation should be minimized to reduce the influence of the non-ablation area on polarization. The influence of the main laser parameters on ablation morphology is investigated. To verify the ablation results, optical microscopy and environmental scanning electron microscopy are used to characterize the surface morphology, and a spectrophotometer is used to test the transmittance. The polarization filter is affixed to a camera for the verification of polarization imaging. The influence of optical crosstalk is analyzed using the camera aperture. The performance of the polarizer is verified by the polarizing film covering or semi-convering camera lens.
The total laser energy projected onto the surface of the polarizer must be strictly controlled during laser processing. Three main laser processing parameters are identified: the laser power density, pulse overlapping ratio, and scan line spacing. During the experiments (Fig. 3), the uniformity of the ablation morphology is mainly determined by the laser pulse overlapping ratio and scanning line spacing. The laser power density determines the energy that a single laser pulse projects onto the polarizer and affects the ablation area, thereby affecting the ablation uniformity. Simultaneously, these three parameters also have coupling effects on the laser ablation process. The main laser ablation parameters are experimentally optimized with an optimized laser power density of 5.89×105 W/cm2, a pulse overlapping ratio of 82.17 %, a scanning line spacing of 0.008 mm, and a laser scanning ablation speed of 3000 mm/s. The morphological characterization (Fig. 4) of the laser-ablated polarizers shows that there are still some grid structures and residues in the ablated area. Removing these residues without damaging the substrate is difficult. According to experimental observations and light transmission tests (Fig. 5), these residues show almost no influence on the transmittance and polarization properties. There are no polarization properties in the ablation area, and the light transmission of the ablation area is 0.96, which is high enough to meet the requirements of the polarization array.
In this study, grating array polarization filters with pixel sizes of 100 μm and 200 μm are fabricated (Fig. 6). An imaging test is performed for the prepared polarizer, and the negative effects of the extinction ratio and optical crosstalk are analyzed and discussed (Fig. 7). The test results demonstrate that the polarization calculated by this polarization filter is accurate, and the angle error between the calculation and actual value is less than 0.5° (Fig. 8).
In the present study, a novel fabrication process for polarization filters is proposed for polarization imaging using the DoFP method. A picosecond laser is used to ablate the subwavelength metal grating polarizer, which eliminates polarization in the ablated area while retaining polarization in the non-ablated area to form a regular polarization array. The 0°, 45°, and 90° polarization arrays are fitted together by cementation to obtain a polarization filter with three-directional polarization channels. In the laser ablation area, the polarization properties diminish, and the light transmittance becomes greater than 0.96. The prepared polarizer retains the high extinction ratio of commercial polarizers and reduces optical interference and pixel registration. The polarization imaging results show that the polarization angle error calculated using the polarization filter is less than 0.5°, and the polarization state can be accurately identified. In summary, the polarization filter has good application prospects in the field of image recognition and polarization imaging.
