INFRARED
Co-Editors-in-Chief
Guilin Chen
Peng ZHANG, Yi MA, and Yuan NIE

The indium deposition situation and microcosmic structure in the indium bump photoresist hole are analyzed in the indium bump growth process of infrared detectors with different pixel pitches. The reason for the low indium bump under the condition of 10 m pixel pitch is explained. And the relationship between the pixel pitch and indium bump height after lift-off is given. For the indium bump growth of infrared detectors with a smaller pixel pitch of 10 m and below, the methods to solve the problem of indium bump height are presented. After using the new methods, the indium bump height can reach more than 5 m.

Jan. 01, 1900
  • Vol. 44 Issue 7 1 (2023)
  • Chang BEN, Ye-xuan DU, Jia-liang CHUANG, and Bai-jian YANG

    The relationship between the electromagnetic shielding characteristics and the infrared transmittance of the metal grid is analyzed through theoretical calculation. It is found that the electrical characteristics of the metal grid are contrary to the optical characteristics. In order to ensure that the transmittance of the optical lens decreases by less than or equal to 5% and the shielding effectiveness is greater than 30 dB (frequency range: 14 kHz-18 GHz) after metal grid plating, it was first found through theoretical calculation that selecting a metal mesh (Ni-Ag) with a line width of 10 m, a period unit length and width of 500 m, and a film thickness of 1 m can meet the requirements. Then, using DC magnetron sputtering technology, the surface of the Zinc sulfide substrate with a diameter of 100 mm was filmed. After exposure, development, hardening, corrosion, stripping and other processes, the infrared window sample with Ni-Ag frequency selective surface(FSS) grid plating was finally obtained. Finally, the infrared window sample was tested for transmittance and shielding effectiveness. The test results met the design requirements. The calculated data can be used to predict the electromagnetic shielding performance of the optical window and provide a reference for the subsequent design work.

    Jan. 01, 1900
  • Vol. 44 Issue 7 8 (2023)
  • Tian-yang LI, and Xing-jie CHEN

    In order to eliminate the non-uniformity of infrared images on uncooled infrared focal plane array and the blind elements generated in the imaging process, the image contrast and the real-time processing ability of infrared images are improved. In this paper, an electronic experimental platform based on the 120×90 infrared focal plane detector produced by Wuhan Guide Infrared company is built. Using the parallel processing and pipeline operation capabilities of field-programmable gate array (FPGA), two-point correction in non-uniformity correction, blind element compensation based on median filter, histogram equalization and false color transformation are realized on FPGA. The experimental results show that under the condition of 200 MHz input clock, this infrared image real-time processing module takes 604 s to process a frame of image, and has strong real-time performance. The research has practical significance for the application optimization of uncooled infrared focal plane arrays.

    Jan. 01, 1900
  • Vol. 44 Issue 7 15 (2023)
  • Jin XU

    Under the influence of Moore's law, the linewidth size of semiconductor manufacturing gradually reaches its limit. In the current process of 28 nm and below, the polycrystalline silicon gate etching is generally adopted by double-layer linked hard mask etching and polycrystalline silicon etching, which achieve effective control of key dimensions and increase the incidence of particle defects. In this paper, the source and formation mechanism of stick particle defects in the process of hard mask etching of polysilicon are analyzed. The removal of particles in the etching cavity and the electrostatic adsorption of the crystal backside are enhanced by precisely adjusting the release time of electrostatic-chuck (ESC) on the wafer and the release time of its own charge after the etching. The results show that when the release time of the wafer is increased by 2 s and the release time of the ESC charge is increased by 6 s, the stick particle defects can be reduced by about 80%. By controlling the associated process parameters to reduce defects, the use of expendable parts and the production costs can be effectively reduced.参考文献原文>Meng X G, Lu L, Chen F H, et al. Challenges and Solutions of 28nm Poly Etching \[C\]. Shanghai: Semiconductor Technology International Conference, 2017.

    Jan. 01, 1900
  • Vol. 44 Issue 7 26 (2023)
  • Yong-zhuang ZHANG, Zhi-hong GONG, Qi-chao RAO, Xu-chen LU, Peng-lei HAN, Jian-guo WEN, and Li-hong GENG

    The heat dissipation problem of the Stirling refrigerator driver components has always been the main limiting factor for the stable operation of the Stirling refrigerator in the high-temperature environment. Taking a certain Stirling refrigerator as the research object, its heat dissipation characteristics are optimized by designing the heat dissipation structure of specific driver components. The reliability of the optimized heat dissipation structure and the heat dissipation improvement effect are verified through experiments. Under the condition of a thermal vacuum environment temperature of 60℃, the average temperature drop of the driver components of the Stirling refrigerator using the optimized heat dissipation structure is about 718%. According to the experiments, it is found that under the condition of a thermal vacuum environment temperature of 60℃, the average temperature of components using red copper as the backplane material is about 905% lower than that of components using aluminum as the backplane material, and the average temperature of components when the refrigerator is placed in the reverse direction is about 1925% lower than that when it is placed in the forward direction. The designed heat dissipation optimization structure, experimental process and results in this paper have certain guiding significance for specific engineering practice .

    Jan. 01, 1900
  • Vol. 44 Issue 7 34 (2023)
  • Gen WANG, Song YUAN, Song YE, Feng XIE, and Jiao CHEN

    Wind fields are crucial to the evolution and prediction of weather situations. Based on the medium wave channel data of GIIRS and the wind field data of ERA5, LightGBM is used to retrieve the three-dimensional atmospheric wind field in this paper. First, model feature variables are constructed. The two-step feature selection method is adopted for the optimal selection of GIIRS channels: (1) The blacklist of GIIRS channels is established; (2) Feature variables are selected by PFI method, and feature variables containing spatiotemporal information are constructed on the basis of forming optimal subsets of channels. Secondly, a three-dimensional wind field retrieval method based on LightGBM is constructed. Finally, LightGBM hyperparameter optimization and correlation retrieval experiments are carried out based on GIIRS encrypted data during Typhoon “Lekima”. The experimental results show that RMSE of the U and V components of the wind field in the test set is less than 1 m/s and 15 m/s respectively, compared with the ERA5 wind field data. The two-step feature selection method in this paper can realize the dynamic optimal selection of GIIRS channels.

    Jan. 01, 1900
  • Vol. 44 Issue 7 39 (2023)
  • Meng XIE, Chun-lei YANG, Ming-jian GU, and Yong HU

    The image resolutions of some channels of Fengyun-4A (FY-4A) satellite are low, which affects the ability of multi-channel collaborative monitoring of Fengyun satellites. To solve this problem, enhanced super-resolution generative adversarial networks (ESRGAN) algorithm is proposed to implement the super-resolution reconstruction of FY-4A images. Based on the generator architecture of ESRGAN, the transfer learning strategy is used, and the pre-training weight is taken as the initial value of the model in this method. A set of residual-in-residual dense blocks (RRDB) containing 5 hollow convolutional layers is designed, and the loss function is optimized. The results show that under the reconstruction of 4 times image resolution, compared with ESRGAN algorithm, the peak signal-to-noise ratio (PSNR), structural similarity (SSIM) and correlation coefficient (CC) of the improved ESRGAN algorithm increase by 0704, 0029 and 0002 respectively, while root-mean-square error (RMSE) decreases by 10%. The reconstructed images are clearer and more natural, and the texture is more detailed.

    Jan. 01, 1900
  • Vol. 44 Issue 7 46 (2023)
  • Please enter the answer below before you can view the full text.
    8+8=
    Submit