【作者】 刘晓鹏；

【导师】 李庆忠；

【作者基本信息】 江南大学 ， 机械设计及理论， 2011， 硕士

【摘要】 化学机械抛光(chemical mechanical polishing，cMP)是现今加工硬脆材料最实用的方法，在微电子和光电了产业应用极其广泛。自从上世纪80年代首次由IBM公司在集成电路芯片抛光应用以来，经过不断探索和机理研究，应用方法已经趋于完善。但传统化学机械抛光技术存在抛光液利用率低、工艺质量不好控制，同时造成环境污染较大。针对上述问题和以材料精细化去除为目的，在中国博士后基金和江苏省自然科学基金的资助下，提出了超声波精细雾化化学机械抛光方法。超声波精细雾化化学机械抛光方法的原理是对特种抛光液的组分进行控频超声精细雾化，形成索太尔直径为5～15um的均匀微米级液粒，通过特殊方式导入到抛光／研抛界面，经过一系列化学和机械作用之后，最后得到光滑无损伤超精纳米级表面。本文介绍了超声精细雾化抛光实验系统工艺流程的确定、实验设备的配置、抛光实验系统的设计及关键技术的解决方案。对抛光机进行了密封改造，防止设备内部被腐蚀；用有机玻璃罩将抛光反应密封在一定的空问内；在玻璃罩E搭建支架，采用极坐标实现对抛光雾液喷嘴的三维精确定位：利用负压方案将抛光雾液导入到抛光界面，实现抛光雾液均匀的分散和有效地利用；用分流的方式实现了去离子水流量的调节，用添加压力块的方式对压力进行调节；采用两级去离子水过滤的方式对抛光液进行过滤和回收。本文对喷嘴的三维坐标和压力进行了正交试验，试验结果表明：超声精细雾化化学机械抛光实验系统能够实现抛光液的超声精细雾化。超声精细雾化化学机械抛光后的表面相糙度可以达到传统化学机械抛光表面l卡日糙度的数量级，但抛光液的用量仅为传统化学机械抛光的1／10左右。本次正交试验的优化参数组合为：压力10Psi、喷嘴高度5mm、雾化半径30mm、喷嘴角度一30。，此时材料的去除率达到最大值113．734nm／min。在选取的四个冈素中，雾化半径对材料去除率的影响最显著，其次为喷嘴角度。通过对抛光机转速、压力、喷嘴高度、雾化半径和喷嘴角度等五个参数进行的单因素试验，得出如下结论：材料去除率随着抛光机转速和压力的增大而增大，随着喷嘴高度、雾化半径和喷嘴角度的增大而减小。通过CSPM5000扫描探针显微镜系统对试件表面的观测，表面粗糙度均在10nm以下，都达到了传统化学机械抛光表面粗糙度的数量级。试验得到的表面粗糙度变化不明显，可以认为在本文做的超声精细雾化化学机械抛光试验中，机械作用占主要地位，所用抛光液中磨粒粒径偏大也是影响表面粗糙度的一个重要因素。更多还原

【Abstract】 Chemical mechanical polishing (CMP) is now the most practical method of brittle materials, and widely used in microelectronics and optoelectronics. Since the first application in integrated circuit chip by IBM in the 80s of last century, it has been perfected through constant exploration and mechanism research. However traditional CMP is in low utilization factor and not easy to control. It also cause environment pollution. In response to these problems and for the purpose of fine material remove, the method of ultrasonic subtle atomization—chemical mechanical polishing is proposed with the help of China Postdoctoral Science Foundation and Natural Science Foundation of Jiangsu Province.The method of chemical mechanical polishing using slurry which is ultrasonic subtle atomized is proposed to use specific frequency ultrasonic to atomize special slurry and get uniform micron-sized droplets whose diameters are between 5u.m and 15um. The droplets are imported to polished/polishing interface by special way, and finally the specimen gets the surface which is super-smooth in nanometer without damage after a series of chemical and mechanical reactions.This article describes the confirmation of the experimental system’s process, the laboratory equipments’ placement, the design of the polishing experimental system and the solution of key technologies. The polishing machine is transformed in order to prevent it from corrosion. Organic glass is used to seal the polish reaction in a certain space. Frame is built on the glass, and polar is used to achieve the three dimensional precise location of the nozzle. Negative pressure is to import the droplets to the polishing interface, so the droplets can be uniform distributed and effective used. Shunt is used to adjust the deionized water flow. The pressure is adjusted by the pressure piece. Slurry is filtered and recycled by two grade filtration equipment.In this paper, pressure and three-dimensional coordinates of the nozzle were orthogonal tested, and the result shows: the ultrasonic subtle atomization experiment system can realize ultrasonic subtle atomization. Surface roughness after ultrasonic subtle atomization CMP is in the same grade as the surface roughness after traditional CMP, but amount of slurry in ultrasonic subtle atomization CMP was only about 1/10 compared to traditional CMP. In this experiment the maximum of MMR was 113.734nm/min while pressure was 10Psi, nozzle height was 5mm, the distance between polish pad center and nozzle center was 30mm and nozzle angle was -30°. The distance between polish pad center and nozzle center has the most significant effect to MMR in the four selected parameters.Five parameters which are polishing machine speed, nozzle height, pressure, nozzle angle and nozzle radius are researched alone. It is concluded that: the material removal rate increases with polishing machine speed and the pressure while the nozzle height, nozzle radius and nozzle angle decreases. The specimen surface is observed by CSPM5000 scanning probe microscope system. Surface roughness are all below lOnm, and have reached the magnitude of the surface roughness in traditional CMP. There is no obvious change in surface roughness. So it can considered the mechanical action is dominant in the ultrasonic subtle atomization CMP experiments. The large size of particle in the slurry is an important factor to influence surface roughness.更多还原