-
复合型高深宽比沟槽标准样板主要用于校准硅基MEMS高深宽比结构的无损测量系统。为了能够准确校准该系统,标准样板应具备复现高深宽比量值的功能,并且为了提高测量的准确性,同时考虑到无损测量系统中包含显微成像的功能,标准样板还需要具备准确定位测量位置及测量角度和辅助定值溯源样板的功能。为了实现标准样板的应用目的,设计了以下特征结构:具有深度、宽度尺寸的深槽(台阶)结构,复现MEMS工艺中需要测量的各种沟槽结构;沟槽辅助定值结构,通过该结构对样板进行剖开,准确复现其高深宽比量值,完成溯源;测量定位结构,准确定位标准沟槽所测量的位置;定位角结构,准确确定样板的测量方向。
根据样板的使用要求,设计的样板图如图1所示,其中高深宽比沟槽结构的设计尺寸为:宽度为标准沟槽宽度、加工刻深尺寸为沟槽深度,长度为2 mm,占于中间位置;测量定位结构设计尺寸为:宽度与标准沟槽宽度一致,取上中下3个位置,每两段之间的空隙用来准确定位测量结构位置,其宽度与标准沟槽宽度一致,辅助定值结构用于确定样板划片位置和划片厚度,厚度尺寸设计为0.5 mm,方便在电镜类仪器下测量,定位角结构为4个具有方向性的小三角,分别位于样板的4个角,成对匹配,辅助测量时准确找到标准样板结构。
表1给出了标准样板的设计尺寸。
表 1 设计标准样板的标称尺寸及深宽比
Table 1. Nominal size and aspect ratio of design standard template
Nominal width/μm Nominal depth/μm Aspect ratio 30 300 10∶1 30 200 Near 7∶1 2 60 30∶1 2 10 5∶1 测量中先对辅助定值结构进行切片,获得高深宽比沟槽结构的截面;使用扫描电镜或原子力显微镜等相应仪器对截面尺寸进行测量得到辅助定值结构的宽度和深度尺寸;将其赋值给高深宽比沟槽结构。如此赋值源于当前的计量方法无法直接对高深宽比达到30∶1的标准样板进行直接表征,但是切片后样板的截面结构可通过电镜类仪器进行直接测量。
-
采用半导体工艺制备标准样板,考虑到标准样板结构与功能的特殊性,常规的工艺制备流程无法完成。为了实现样板设计中的各种结构与特殊功能,需要有针对性的设计工艺流程。图2是针对复合型高深宽比沟槽标准样板进行的工艺设计流程图,包括SOI片的制备、光刻、刻蚀、氧化以及腐蚀等,关键工艺为光刻和刻蚀。
图 2 高深宽比沟槽标准样板的工艺设计流程图
Figure 2. Process design flow chart of high-aspect-ratio trench standard template
(1) 材料选择与制备:标准样板的选材方面不仅需要考虑加工的工艺特性,更需要考虑样板的加工质量,以实现沟槽宽度和高度分布的高度一致性。研制的硅基标准样板当台阶高度尺寸大于20 μm时加工的质量很差,底面明显不平整,导致测量误差很大[11]。参考文献[12]的设计,提出采用绝缘衬底上的硅片,即Silicon-On-Insulator (SOI),SOI片作为加工基板。采用双面抛光的硅晶圆片作为衬底材料,清洗干净,表面生长一层二氧化硅[13],然后采用键合的方法使其与另一片清洗干净的硅晶圆片结合组装成SOI片,其加工层的硅片厚度依据所用要求进行研磨得到,具体流程图如图3所示。
(2) 关键工艺:光刻和刻蚀。为了提高样板的质量采用投影光刻和干法刻蚀完成样板的结构加工过程,投影光刻的特点为分辨率高,重复性好,不易损坏掩膜版。投影光刻的最小分辨力为0.15 μm,加工的所有样片的尺寸均可以采用该工艺完成。干法刻蚀其突出优点为可以获得及其精确的特征图形。
(3) 平整度处理:由于样板的高深宽比参数大于等于10∶1,首次刻蚀后的样板沟槽侧壁会存在很多毛刺,侧壁的粗糙度较大会给测量造成很大误差。采用氧化工艺对其表面生长一层比较薄的二氧化硅,然后再进行二次腐蚀,将二氧化硅层去除,提高了样板沟槽侧壁的平整度[14]。
图4所示为基于以上工艺研制出的复合型高深宽比标准样板。
-
复合型高深宽比沟槽标准样板可以用于校准近红外宽光谱干涉显微测量系统。近红外宽光谱干涉显微技术的基本原理是基于干涉测量法。硅材料在可见光波段不透明,但在近红外波段的透过率较高。采用近红外的宽带光源,穿透硅基MEMS的高深宽比结构,可获取宽带干涉光谱信号,进而提取高深宽比结构的形貌信息。大NA近红外宽光谱干涉显微无损测量系统可基于透射、反射模态实现高深宽比硅基MEMS器件的测量[18-19]。
使用近红外宽光谱干涉显微测量系统对上述表征后的复合型高深宽比沟槽标准样板进行测试。测试过程中首先通过定位角结构找准测试方向,然后再扫描正交扫描标定结构确定测量角度,最后通过测量定位结构选定测量位置,开始测试。测试结果如表2所示。对于线宽2 μm、沟槽深度10 μm的样板其线宽尺寸的平均值为1.8 μm,与定值结果的偏差为0.03 μm,沟槽深度尺寸的平均值为9.9 μm,与定值结果的偏差为0.09 μm;对于线宽2 μm、沟槽深度60 μm的样板其线宽尺寸的平均值为2.3 μm,与定值结果的偏差为0.10 μm,沟槽深度尺寸的平均值为60.2 μm,与定值结果的偏差为0.40 μm;对于线宽30 μm、沟槽深度200 μm的样板其线宽尺寸的平均值为30.7 μm,与定值结果的偏差为1.10 μm,沟槽深度尺寸的平均值为200.7 μm,与定值结果的偏差为0.40 μm。对于线宽30 μm、沟槽深度300 μm的样板其线宽尺寸的平均值为30.8 μm,与定值结果的偏差为1.20 μm,沟槽深度尺寸的平均值为300.4 μm,与定值结果的偏差为0.80 μm。以上数据显示测试结果与样板的SEM表征结果基本一致。图13给出了线宽30 μm、沟槽深度300 μm的样板在近红外宽光谱干涉显微测量系统下的测试结果图。
表 2 样板定值结果与仪器测量结果对比
Table 2. Comparison of sample template setting results and measurement results
Model standard
value/μmCharacterization
results/μmMeasurement
results/μmDeviation/μm Line width 2 1.83 1.8 0.03 Trench depth 10 9.99 9.9 0.09 Line width 2 2.20 2.3 0.10 Trench depth 60 59.80 60.2 0.40 Line width 30 29.60 30.7 1.10 Trench depth 200 200.30 200.7 0.40 Line width 30 29.60 30.8 1.20 Trench depth 300 299.60 300.4 0.80
A hybrid high-aspect-ratio trench standard template
-
摘要: 硅基MEMS器件中存在大量高深宽比结构,对这些结构进行线宽和深度的无损检测,是当前的热点问题。为了实现对这些高深宽比结构无损测量系统的准确校准,采用半导体工艺研制了一系列高深宽比沟槽标准样板,宽度范围2~30 μm、深度范围10~300 μm,其深宽比最大达到30∶1。为了满足样板的校准功能,设计了多种特征结构,包括辅助定值结构、测量定位结构和定位角结构等,还设计了样板量值的表征与考核方法。考核量值包括线宽尺寸、沟槽深度尺寸和均匀性。使用扫描电镜对标准样板进行了测试,结果表明该标准样板可以用于校准近红外宽光谱干涉显微测量系统。Abstract:
Objective There are a large number of high-aspect-ratio structures in silicon-based MEMS devices, and non-destructive testing of linewidth and depth of these structures is a hot issue at present. Generally, the depth-to-width ratio of MEMS high-aspect-ratio structures is generally between 10∶1 and 100∶1, and the trench width is a few microns to tens of microns. At present, in the silicon-based MEMS process line, anatomical testing is the main means of high-aspect-ratio structure testing, but there are the following defects: it is necessary to use scanning electron microscopy (SEM) for auxiliary measurement, which is inefficient and cumbersome; It is a destructive measurement that causes irreversible damage to MEMS products; It can only be sampled and cannot fully reflect the characteristics of the process. Based on this, a non-destructive measurement system with high-aspect-ratio structure near-infrared broad-spectrum microscopy measurement system was developed, and its measurement accuracy will directly affect the overall performance of the device under test, so it is necessary and urgent to calibrate the measurement system. Methods In order to achieve the accurate calibration of the non-destructive measurement system of high-aspect-ratio structure, a series of standard samples of high-aspect-ratio trenches are designed and developed by semiconductor process, with a width range of 2-30 μm, a depth range of 10-100 μm, and a maximum high-aspect-ratio of 30∶1 (Tab.1). The samples were characterized and fixed, and finally the developed standard samples were applied to the calibration of the near-infrared broad-spectrum microscopy measurement system (Fig.13). Results and Discussions In order to meet the calibration function of the standard samples, a variety of characteristic structures are designed (Fig.1), including auxiliary fixed value structure, measurement positioning structure and positioning angle structure, etc., and the characterization and assessment method of the sample value are designed (Fig.5-6). Measurement values include line width size, trench depth size, and uniformity. Finally, the developed standard template is applied to the near-infrared broad-spectrum microscopy measurement system to further verify the accuracy of the developed system, that is, the applicability of the template (Tab.2). Conclusions In order to solve the calibration problem of the near-infrared broad-spectrum interferometric microscopy system, a series of standard samples of high-aspect-ratio grooves were developed, with a width range of 2-30 μm and a depth range of 10-300 μm, and its high-aspect-ratio reached a maximum of 30∶1. In order to meet the calibration function of the template, a variety of characteristic structures are designed, including auxiliary fixed value structure, measurement positioning structure and positioning angle structure, etc., and the characterization and assessment method of the sample measurement value is designed. Since there is no suitable measuring instrument to directly characterize the value of the standard template of the composite high-aspect-ratio trench, an auxiliary fixed value structure is designed for the standard template, and the cross-section of the high-aspect-ratio trench structure is displayed by sectioning, and then it is measured by scanning electron microscope or atomic force microscope, and the uniformity of the template is characterized to ensure the consistency of the measurement results of the template. Finally, the developed standard template was measured by the near-infrared broad-spectrum interferometric micrometry system, and the measurement results showed that the magnitude was basically consistent with the characterization results. -
Key words:
- high-aspect-ratio /
- line width /
- trench depth /
- trench standard template
-
表 1 设计标准样板的标称尺寸及深宽比
Table 1. Nominal size and aspect ratio of design standard template
Nominal width/μm Nominal depth/μm Aspect ratio 30 300 10∶1 30 200 Near 7∶1 2 60 30∶1 2 10 5∶1 表 2 样板定值结果与仪器测量结果对比
Table 2. Comparison of sample template setting results and measurement results
Model standard
value/μmCharacterization
results/μmMeasurement
results/μmDeviation/μm Line width 2 1.83 1.8 0.03 Trench depth 10 9.99 9.9 0.09 Line width 2 2.20 2.3 0.10 Trench depth 60 59.80 60.2 0.40 Line width 30 29.60 30.7 1.10 Trench depth 200 200.30 200.7 0.40 Line width 30 29.60 30.8 1.20 Trench depth 300 299.60 300.4 0.80 -
[1] Gabriel K J. Micro electro mechanical systems [C]//Proceedings of the IEEE, 1998, 86 (8): 420-457. [2] Greywall D S. MEMS safety and arming devices having launch and rotation interlocks and method of manufacturing the same: US, US7371982 B2[P]. 2008-05-13. [3] Grace R H. Application opportunities of mems/mst in the automotive market: The great migration from electromechanical and discrete solutions [J]. Advanced Microsystems for Automotive Applications, 2001, 17(3): 1-16. [4] Jo T, Kim S, Pahk H. 3D measurement of TSVs using low numerical aperture white-light scanning interferometry [J]. Journal of the Optical Society of Korea, 2013, 17(4): 317-322. doi: 10.3807/JOSK.2013.17.4.317 [5] Hyun C, Kim S, Pahk H. Methods to measure the critical dimension of the bottoms of through-silicon vias using white-light scanning interferometry [J]. Journal of the Optical Society of Korea, 2014, 18(5): 531-537. doi: 10.3807/JOSK.2014.18.5.531 [6] Peng B, Hou W, Xu Q. Precision 3D profile in-line measurement of through-silicon via (TSV) based on high-frequency spectrum signals in the pupil plane [J]. Optics Communications, 2018, 424: 107-112. doi: 10.1016/j.optcom.2018.04.033 [7] Krauter J, Osten W. Nondestructive surface profiling of hidden MEMS using an infrared low-coherence interferometric microscope [J]. Surface Topography: Metrology and Properties, 2018, 6(1): 015001. [8] Shi J H, Han B C. Internal profile reconstruction of microstructures based on near-infrared light transmission reflection interferometry with optical path compensation [J]. Chinese Optics, 2019, 12: 395-404. doi: 10.3788/co.20191202.0395 [9] Bennett J M. Comparison of instruments for measuring step heights and surface profiles [J]. Applied Optics, 1985, 24(22): 3766-3772. doi: 10.1364/AO.24.003766 [10] Potzick J E, Nunn J W. International comparison of photomask linewidth standards: United States (NIST) and united kingdom (NPL)[C]//Proceedings of SPIE, 1996, 2725: 124-129. [11] Feng Yanan, Li Suoyin, Han Zhiguo, et al. Selection of the etching processes for the fabrication of the micro step height specimens [J]. Micronanoelectronic Technology, 2016, 53(11): 773-778. (in Chinese) doi: 10.13250/j.cnki.wndz.2016.11.011 [12] Seah M P. CCQM-K32 key comparison and P84 pilot study: Amount of silicon oxide as a thickness of SiO2 on Si [J]. Metrologia, 2008, 45(1A): 08013. doi: 10.1088/0026-1394/45/1A/08013 [13] Gore S P, Funde A M, Salve T S, et al. Properties of silicon dioxide films prepared using silane and oxygen feeds by PE-CVD at low power plasma [J]. Journal of Nano- and Electronic Physics, 2011, 3(1): 370-375. [14] Hanssen L M, Wei H C, Lo C W, et al. The design of microscope type spectral reflectometry for the depth measurement of high-aspect-ratio through silicon via[C]//Proceedings of SPIE, 2016, 99610: 99610L. [15] Lei Lihua, Zhang Xinyin, Wu Junjie, et al. Characterization of nanofilm parameters based on hybrid optimization algorithm [J]. Infrared and Laser Engineering, 2020, 49(2): 0213001. (in Chinese) doi: 10.3788/IRLA202049.0213001 [16] UNE-EN ISO 25178-70-2015. Geometrical product specification (GPS) - Surface texture: Areal - Part 70: Material measures (ISO 25178-70: 2014)[S]. Austria: Osterreichisches Normungsinstitut, 2015. [17] Bodermann B, Zurbrick L S, Montgomery M W, et al. Results of an international photomask linewidth comparison of NIST and PTB[C]//Proceedings of SPIE, 2009, 7488: 74881H. [18] Kessel T V, Wickramasinghe H K. Measurement of trench depth by infrared interferometry [J]. Optics Letters, 1999, 24(23): 1702-1704. doi: 10.1364/OL.24.001702 [19] Lebrasseur E, Pourciel J, Bourouina T, et al. A new characterization tool for vertical profile measurement of high-aspect-ratio microstructures [J]. Journal of Micromechanics and Microengineering, 2002, 12(3): 280-285. doi: 10.1088/0960-1317/12/3/313