Polishing Characteristics of Heavily Doped Silicon Wafers with Different Dopants

被引:0
|
作者
Zhou Y. [1 ]
Zhang G. [2 ]
Zhou Q. [1 ,2 ]
Shi X. [2 ]
Lin L. [2 ]
Lu Y. [2 ]
机构
[1] General Research Institute for Nonferrous Metals, Beijing
[2] GRINM Semiconductor Materials Co., Ltd., Beijing
来源
关键词
Chemical mechanical polishing; Dopant; Heavy doped silicon;
D O I
10.13373/j.cnki.cjrm.XY19010034
中图分类号
学科分类号
摘要
Nowadays, with the rapid development of information technology, the requirements for the manufacturing of large-scale integrated circuits (IC) on the substrate silicon wafer were becoming important. In order to improve the overall performance of IC, wafer processing technology in the silicon wafer surface state, microscopic flatness and other aspects of the requirements had reached in the nanometer level. Chemical mechanical polishing (CMP) was the final leveling process for the surface of the silicon wafer in the processing link of the substrate. In the polishing process, the surface of the silicon wafer would be subjected to the grinding effect of the micro particles in the slurry and the corrosion effect of the slurry at the same time. Under the alternating action of mechanical and chemical aspects, the ultra-precision flat surface processing was realized. However, many parameter variables in CMP technology, such as polishing fluid, polishing pressure, polishing speed and polishing time, would affect the polishing effect. Therefore, in this paper, CMP polishing experiments were carried out on three kinds of heavily doped silicon wafers with different dopants to analyze the influence of different dopants on the polishing removal rate and the surface micro-roughness after polishing. A single-sided CMP polishing machine was used to polish silicon wafers and 54 silicon wafers doped with boron, arsenic and antimony with crystal direction <100> and diameter of 150 mm were selected. KOH polishing solution and SiO2 abrasive were used. Each silicon wafer had 18 pieces. In the polishing process, the pressure head pressurized the ceramic plate, and then the silicon wafer was pressurized on the rotating polishing pad. At the same time, the polishing fluid was added to make the mechanical action and chemical action took place simultaneously. Each group of wafers went through three polishing processes, in which the roughing time was 8 min. After finishing the polishing, all the silicon wafers were cleaned by the cleaning tank, and the residual polishing fluid on the surface was removed for testing and analysis. This experiment focused on the influence of different dopant types on the polishing removal rate and the difference of the surface roughness and surface state of silicon wafers with different dopant types after full polishing. ADE9600 equipment was used to measure the thickness of silicon wafer before and after polishing, and the polishing removal rate of silicon wafer was calculated. The Haze value of the polished wafer was detected by KAL Tensor SP1 particle detection equipment. Two pieces of each silicon wafer were randomly selected, and the surface roughness and surface morphology were measured by white light interferometer. Analyzed the experimental data and got the results that the polishing removal rate of heavily boron-doped silicon wafer was the smallest, and the polishing removal rate of heavily arsenic-doped silicon wafer was the highest. After polishing, Haze value of heavily boron-doped silicon wafer was the largest, while Haze value of heavily arsenic-doped and heavily antimony-doped silicon wafer was small. It indirectly reflected that the surface micro-roughness of heavily boron-doped silicon wafer was higher than that of heavily arsenic-doped silicon wafer and heavily antimony-doped silicon wafer as a whole. White light interference test displayed that the surface micro-roughness center of heavily boron-doped silicon wafer was low and the edge was high, and the heavily arsenic-doped silicon wafer was obviously higher than that of the center and the heavily doped antimony silicon wafer was consistent from the center to the edge. The conclusions of this experiment were as follows: (1) The polishing removal rate of heavy boron doped silicon was obviously lower than that of heavy arsenic doped and heavy antimony doped silicon. (2) The polishing removal rate of heavy arsenic doped silicon wafer was higher than that of heavy antimony and heavy boron doped silicon wafer. (3) The overall surface roughness of the heavy boron doped silicon wafer was the largest, and the surface morphology of the heavy boron doped silicon wafer was low in the center and high in the edge. (4) The overall surface roughness of the heavy arsenic doped silicon wafer was small, and the surface morphology of the heavy arsenic doped silicon wafer was low in the center and high in the edge. (5) The surface roughness of heavy antimony doped silicon wafer was small, and the surface morphology was relatively smooth. In this study, the experimental variables could be further carefully controlled and the influence of uncertain factors in polishing could be reduced by a large number of repeated experiments. By analyzing the surface state of different silicon wafers after polishing, the process parameters could be obtained to improve the polishing effect of different silicon wafers, which was conducive to improving the product quality in the actual production. © Editorial Office of Chinese Journal of Rare Metals. All right reserved.
引用
收藏
页码:1018 / 1024
页数:6
相关论文
共 20 条
  • [1] Lei H, Luo J B, Zhang C H., Research progress of chemical mechanical polishing technology, Journal of Shanghai university (Natural Science Edition), 6, (2003)
  • [2] Schuler J., CMP technology and markets, SEM I2CON China 99 Techanical Symposium [C], (1999)
  • [3] Liu Y L., Study on CMP polishing technology of silicon wafer, Electronics Process Technology, 31, 5, (2010)
  • [4] Song X L, Liu H Y, Yang H P, Zhang X W, Xu D Y, Qiu G Z., Chemical-mechanical polishing rate and polishing mechanism of semiconductor silicon wafers in nano-SiO<sub>2</sub> slurry, Journal of the Chinese Ceramic Society, 8, (2008)
  • [5] Han B S, Zarudi I., Deformation and material removal rate in polishing silicon wafers, Journal of Materials Processing Technology, 140, (2003)
  • [6] Wu M L, Lin S C., A study of the effects of polishing parameters on material removalrate and non-uniformity, International Journal of Machine Tools & Manufacture: Design, Research and Application, 42, (2002)
  • [7] Zhao Y W, Chang L., A micro-contact and wear model for chemical-mechanical polishing of silicon wafers, Wear, 252, (2002)
  • [8] Keskitalo N, Forsberg M, Olsson J., Effect of dopants on chemical mechanical polishing of silicon, Microelectronic Engineering, 60, (2003)
  • [9] Yang H P., Study on Electrochemistry and Polishing Rate of Semiconductor Silicon Wafer, 1, (2007)
  • [10] Zhong G H, Ning Y D, Wang X, Lu Y C, Zhou Q G, Li Y D., Effect of chemical etching on local flatness of semiconductor silicon wafer after polishing, Chinese Journal of Rare Metals, 42, 11, (2018)