光刻机工件台方镜面形子孔径拼接干涉测量技术研究
文献类型：学位论文
| 作者 | 李永 |
| 学位类别 | 硕士 |
| 答辩日期 | 2015 |
| 授予单位 | 中国科学院上海光学精密机械研究所 |
| 导师 | 唐锋 |
| 关键词 | 干涉测量 子孔径拼接 误差累积 高反射率平面 45°平面镜 |
| 其他题名 | Study on sub-aperture stitching interferometry for testing the flatness of lithography wafer stage mirror |
| 中文摘要 | 光刻机是集成电路制造的关键设备，光刻机工件台是决定光刻机产率、套刻精度等核心指标的重要分系统。高端光刻机工件台一般采用双频激光干涉仪与方镜组成反馈定位传感系统，工件台方镜的面形精度影响工件台同步及扫描定位精度，进而影响光刻机套刻精度与产率。因此高端光刻机对工件台方镜的面形加工检测精度提出了很高要求，如在一维全口径大于350mm的区域内，任意小口径（~30mm）面形需优于λ/25 PV，加工过程中小口径面形检测精度需达到~λ/100 PV（λ=632.8nm），检测口径大于350mm；此外还需实现高精度高反射表面面形测量、45°角反射镜测量。目前商用检测仪器尚不能满足其检测精度、检测口径，及高精度高反测量等测试需求。 子孔径拼接技术能够实现高空间分辨率、高子孔径测量精度面形检测。本文针对高端光刻机工件台方镜面形子孔径拼接干涉测量应用，围绕提高子孔径拼接干涉测量系统检测精度、高反射率平面面形高精度检测以及45°平面镜面形子孔径拼接检测开展研究，主要工作如下： (1) 提出了一种变权重数据融合拼接算法，有效消除子孔径拼接缝隙，提高了拼接数据的连续性。基于8个实测子孔径面形数据，以去除36项Zernike多项式拟合面形后的拟合残差、及残差在拼接位置的相位梯度为拼接缝隙评价指标，将变权重数据融合算法与简单平均法相比较，数据连续性提高89倍。 (2) 提出了一种在拼接过程中标定标准镜面形离焦误差，消除拼接累积误差的方法，使得子孔径拼接检测精度与大口径干涉仪相当。针对平面面形子孔径拼接误差累积问题，分析了拼接累积误差产生的原因，推导了参考镜离焦与拼接累积误差、拼接次数间的定量表达式，基于该表达式在拼接过程中标定参考镜离焦误差，降低拼接累积误差。对450mm×60mm的平面镜进行了8个子孔径的拼接检测，与大口径干涉仪检测结果比对，去除参考面离焦误差前后拼接测量误差从λ/10 PV减小至λ/30 PV，有效提高了拼接测量精度。结合绝对检验技术标定参考镜高阶面形误差，验证了离焦是引起拼接误差累积的主要因素，消除参考镜高阶面形误差并不能显著提高拼接检测精度。 (3) 针对工件台方镜高反射平面面形高精度检测需求，优化设计了一种新的Clapham-Dew膜（C-D膜），可实现11%～100%反射率平面面形的高干涉对比度检测。数值仿真分析了高反射率平面面形测量过程中的条纹干涉对比度降低、干涉条纹偏离正弦分布、相位测量误差增大等问题。针对现有商用C-D膜标准镜测量98%反射率平面面形的条纹干涉对比度只有~52%，影响检测精度的问题，推导了斐索干涉仪干涉场与C-D膜反射率、衰减以及相位延迟等参数之间的定量关系，优化提出了一种新的C-D膜技术要求。实验结果表明所设计的C-D膜实现了反射率11%~100%镜面的~80%以上干涉对比度测量。 (4) 提出了一种光刻机工件台方镜45°反射镜面形子孔径拼接检测方案。通过增加反射参考镜，实现了干涉仪与方镜的水平姿态检测，具有操作简便的优点。组合设计了干涉仪参考镜和反射镜的膜系参数，实现镀膜前后45°反射镜面形的高干涉对比度检测，以保证检测精度。 |
| 英文摘要 | Projection lithography tools are critical equipment in the IC manufacture. The wafer stage is one of the most important sub-systems in lithography tool. It determines several kernel specifications of the lithography tool, including throughput, overlay accuracy etc. Usually, double-frequency laser interferometers and wafer stage mirrors are used as the position feedback system of the wafer stage in high-end lithography tool. The surface flatness of the wafer stage mirror has an evident influence on the synchronization, scanning and positioning accuracy of the wafer stage. And then, it will affect the throughput and overlay accuracy of lithography tools. Therefore, high surface flatness specifications are needed for the high-end projection lithography tools. For example, the full aperture of the stage mirror in one direction exceeds 350mm, and its surface flatness should be better than λ/25 PV in every local Φ30mm area. So, the measurement accuracy in local area should be about λ/100（λ = 632.8nm）PV, and the testing aperture should be extended to larger than 350mm. In addition, testing the stage mirror with high reflection coating and placed in 45° angle with high accuracy are also needed for wafer stage mirror measurement. So far, there is no commercial testing equipment which can meet the demands of measurement accuracy and measuring aperture simultaneously, not to mention measuring high reflection surface figure with high accuracy. Sub-aperture stitching interferometry can realize large aperture test with high spatial resolution. And the measurement accuracy of each sub-aperture is high enough for advanced optical manufacturing. Aiming at testing the flatness of high-end lithography wafer stage mirror with sub-aperture stitching interferometer, high precision sub-aperture stitching algorithm for flat mirrors, high reflectance surface flatness test with high visibility and 45 ° flat mirror sub-aperture stitching test are studied in this dissertation. The main content is as follows. (1) A variable weight data fusion stitching algorithm is proposed to remove stitching-gap and improve stitching continuity. Using this algorithm, eight sub-aperture surface figure data are stitched together. The residual fit error of the first 36 Zernike polynomials, and the gradient of the residual error at the stitching position are used to estimate the stitching-gap. Comparing with the result of the simple average method, the stitching continuity is improved by 89 times. (2) A stitching accumulation error reduction method based on calibrating the power of reference mirror in stitching process is proposed. The stitching measurement accuracy can match with large aperture interferometer. For the issue of error accumulation effect in sub-aperture stitching test for flat mirrors, the causes of error accumulation in stitching process are analyzed in detail. A quantitative equation is deduced for calculating the power of reference flat (RF) from stitching accumulation error and stitching numbers. Then the power of RF is calibrated and removed in the process of stitching. The error accumulation is reduced distinctly. A flat mirror with aperture of 450mm×60mm was tested by 8 sub-apertures. Compared with the test result of a large aperture interferometer, the stitching measurement error was reduced from λ/10 PV to λ/30 PV with power compensation. The stitching test accuracy is improved effectively. Using absolute flatness test, the high order surface figure of the RF was also calibrated. It is verified that the power is the main source of the error accumulation in stitching. Removing the high order surface figure of the RF cannot improve the stitching accuracy significantly. (3) For measuring the flatness of coated high reflection lithography wafer stage mirror with high accuracy, a new Clapham-Dew coating film is designed to test 11%～100% reflectance surface with high contrast interference fringes. The problems of low visibility, non-sinusoidal deviation of intensity, and high phase measurement error are analyzed for measuring high reflection surface with regular methods. When using the existing Clapham-Dew coating RF, an interference visibility of only ~52% can be reached. The low interference visibility will affect the measurement accuracy. The interference intensity expression is deduced when the reference surface has Clapham-Dew coat. And a new Clapham-Dew coating specification is designed. The experimental results show that using the new designed Clapham-Dew coating film, the interference visibility is larger than ~80% when the reflectivity of the mirror under test is in the range of 11%~100%. (4) A scheme for testing the 45° lithography wafer stage mirror with sub-aperture interferometer is proposed. By inserting a reflective reference mirror, the test can be easily carried out with both the interferometer and the stage mirror are in horizontal position. The reflectance of reference flat and reflective reference flat are designed together to obtain high visibility for testing lithography wafer stage mirror before and after coating. |
| 语种 | 中文 |
| 源URL | [http://ir.siom.ac.cn/handle/181231/16940]  |
| 专题 | 上海光学精密机械研究所_学位论文 |
| 推荐引用方式 GB/T 7714 | 李永. 光刻机工件台方镜面形子孔径拼接干涉测量技术研究[D]. 中国科学院上海光学精密机械研究所. 2015. |
入库方式： OAI收割
来源：上海光学精密机械研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。