影响牺牲层腐蚀速率的因素研究

精确预测腐蚀速率对于避免过腐蚀和节省时间,从而提高MEMS器件加工的效率具重大意义.通过改变牺牲层材料、腐蚀液浓度、温度和牺牲层结构来改变腐蚀速率是常用的方法.在前人腐蚀模型基础上考虑扩散系数是浓度和温度的函数,腐蚀速率常数是温度的函数,得到了修正模型.从修正模型中找出影响腐蚀速率的各种因素的对应参数,并对其影响腐蚀速率的机理进行详细地研究,这样就为通过修改某些因素来改变腐蚀速率提供了依据.     

组合结构的牺牲层释放腐蚀修正模型

Eaton等人曾给出了HF溶液腐蚀SiO2牺牲层的释放腐蚀模型,然而实验中发现该模型并不能较好地符合实验数据.经分析发现该模型中扩散系数和反应速率常数不能如实反映温度的变化,而且实验中观察到腐蚀前端形状并不总是平面.通过对模型中这些因素的修正,建立起修正模型,证实修正模型比Eaton给出的模型更符合实验结果     

牺牲层腐蚀改进模型与模拟研究

牺牲层腐蚀主要受腐蚀液的扩散过程制约,由扩散方程决定.扩散系数在腐蚀过程中随温度和腐蚀液浓度变化而改变.文中对改进的腐蚀模型给出有限差分算法,由每一时刻溶液在具体位置的浓度值得出扩散系数,再由Topography模型计算前端面的腐蚀情况得到腐蚀前端行进的轮廓线.并编程对一些MEMS结构的释放过程进行仿真,最后给出实验验证.     

电化学选择性刻蚀Cu/Ni牺牲层技术的研究

牺牲层腐蚀技术结合MEMS技术是制作三维可动微机构的一个重要加工手段。采用电位控制的电化学释放牺牲层技术,对2种不同刻蚀液下的Cu/Ni叠层结构分别进行了电化学腐蚀,并测量了其伏安特性,结果表明:电位控制的电化学腐蚀能很好地进行有选择性刻蚀Cu/Ni牺牲层。     

纳米厚度牺牲层腐蚀速率研究

通过大量的实验研究,建立了一套纳米量级牺牲层腐蚀行为的实验研究方法.对牺牲层厚度对腐蚀速率的影响进行了详细地研究,并得到了如下结论:当牺牲层厚度达到微米量级时,其腐蚀速率随着牺牲层厚度的增大而加快,但当其达到纳米量级时,由于固体表面存在的静电荷而产生双电层效应,这种效应对腐蚀速率的影响超过了牺牲层厚度的影响,最终使得腐蚀速率和牺牲层厚度无关.     

牺牲层二维腐蚀修正模型的研究

针对氢氟酸腐蚀氧化硅,深入研究了牺牲层腐蚀的原理。牺牲层腐蚀主要受扩散机制影响。把二维扩散方程中的扩散系数看作溶液浓度和温度的函数,建立了二维腐蚀修正模型。利用有限差分算法求解扩散方程,并使用C语言编程实现了对单开口、内外拐角等多种复杂组合结构腐蚀过程的模拟,使用MATLAB软件绘制腐蚀图形,最后将模拟结果与实验结果进行了对比,验证了模型的合理性。     

纳米梁的金硅原电池腐蚀和无损释放技术研究

采用金电极的硅纳米梁在通过HF湿法腐蚀SiO2牺牲层释放结构的时候会发生硅纳米梁被腐蚀现象,消除此效应对于纳米尺度梁制造非常重要:通过电化学工作站测量不同条件下金/硅在HF中的极化曲线和腐蚀电流,从定性和定量研究此腐蚀的原理和影响因素:金硅在HF中形成的原电池效应是此腐蚀的主要原因;改变金硅面积比和改变HF构成可以减缓...     

CO2激光直写结合湿法刻蚀加工玻璃基微流控芯片

玻璃是制作微流控芯片的重要材料,其加工工艺主要基于光刻后湿法腐蚀,对设备和实验室要求较高.本文提出以普通指甲油和指甲油/金/铬为牺牲层,利用CO2激光烧蚀开窗口,辅以湿法腐蚀加工玻璃基微流控芯片的方法,并考察了激光加工参数,腐蚀液组成,牺牲层等因素对芯片质量的影响.该方法简便易行,不需要光刻的昂贵设备和繁杂步骤.     

用于MEMS的叠层光刻胶牺牲层技术

研究了用于制备悬空结构的叠层光刻胶牺牲层工艺.讨论了工艺中常遇到的烘胶汽泡、龟裂、起皱、刻蚀电镀种子层时产生的絮状物和悬空结构释放时的粘附等问题,并提出了相应的解决办法.借助于分层刻蚀法和逐步替换法,用叠层光刻胶作牺牲层并利用湿法释放技术,制备得到了长1400 μm、厚6 μm、宽40 μm、悬空高为10 μm的完好的悬臂梁结构.     

多层导电结构隐藏腐蚀涡流检测的有限元仿真

研究多层导电结构材料腐蚀检测优化问题,多层结构存在隐蔽性易漏检的难点。针对内部腐蚀缺陷难以有效检测问题,提出利用含隐藏腐蚀缺陷的多层导电结构线圈阻抗变化级数模型求解,获得影响线圈阻抗变化的相关参数。在有限元软件上,对含腐蚀多层导电结构上方线圈的阻抗变化,与一种Maxwell方程组、矢量磁位和空间解域截取的级数展开快速解析方法,进行了仿真计算,并分别对探头线圈激励频率、探头提离高度、多层导电结构表层厚度、探头尺寸等因素进行对比,所得结果吻合,并验证了结果的正确性和有效性,提高检测多层导电结构隐藏腐蚀缺陷的灵敏度,可为多层导电结构复杂缺陷的检测提供理论指导。     

双层布线技术研究

随着集成电路技术的发展,双层布线技术显得越来越重要。主要对双层布线中二次金属淀积前的通孔预处理技术进行了研究和探索,通过对反溅射时间、反溅射功率等不同工艺参数的对比实验,给出了最优的反溅射清洗工艺条件。     

微F-P腔可调谐滤波器关键工艺研究

采用表面加工工艺,AZ5214E光刻胶进行光刻并反转,磁控溅射NiCr合金,剥离出高度为2.3μm的金属桥墩,填充聚酰亚胺作为牺牲层,再在牺牲层上光刻、沉积金属形成金属桥面,在金属桥面的中心嵌入第二布拉格反射镜。采用O2等离子体刻蚀去除聚酰亚胺膜,制作成微法布里—珀罗（ F-P）腔,不需要硅片键合,克服了传统F-P腔高度不够高、调谐范围有限、腔平整度不好以及对设备要求高的缺点,并且可以做出大阵列结构,易于探测器集成。着重对腔体关键工艺,即金属桥墩的NiCr剥离工艺进行研究,针对现有技术缺陷,提出解决办法。     

Germanium as a versatile material for low-temperaturemicromachining

Though germanium (Ge) shares many similar physical properties with silicon (Si), it also possesses unique characteristics that are complementary to those of Si. The advantages of Ge include its compatibility with Si microfabrication, its excellent gas and liquid phase etch selectivity to other materials commonly used in Si micromachining, and its low deposition temperature (<350°C) that potentially allows Ge to be used after the completion of a standard CMOS run. Wider applications of Ge as a structural, sacrificial, and sensor material require a more systematic investigation of its processing and properties. The results of such an undertaking are presently reported. The topics covered are the formation of Ge thin films and novel application of the selective deposition of Ge to etch hole filling, characterization of the effects of thermal treatment on the evolution of the residual stress in Ge thin films, etch selectivity for etch mask and sacrificial layer applications, and gas phase release technique for stiction elimination     

Characterization of High-Pressure $hbox{XeF}_{2}$ Vapor-Phase Silicon Etching for MEMS Processing

Typical release for structures in microelectromechanical systems (MEMS) devices requires the use of sacrificial layers and wet etchants. As an alternative, bulk Si can be utilized for nonsilicon MEMS or structures as the sacrificial material when exposed to vapor-phase XeF₂ . This paper presents the results of using relatively high pressures (> 3.0 torr) for the purpose of MEMS processing, while characterizing the physical etching mechanism and its effects on the working Si substrate in relation to the allowed processing time. The observed etch rates for high-pressure release varied from 1.6 to 1.9 mum/min for applied pressures of 4.5-5.5 torr. The resulting roughness is shown to be primarily dependent on time, where the maximum average roughness is approximately 1.4 mum after 3000 s at 5.5 torr. Slightly anisotropic results are produced by the increased pressures, showing a 0.7 : 1.0 (vertical : lateral) etch rate, as well as some detrimental effects to the released structures. Furthermore, the use of etch windows are investigated in relation to etch rate when subjected to these high pressures.     

Dendritic material as a dry-release sacrificial layer

A dry-release process using highly structured dendritic material, specifically, hyperbranched polymers (HBP's), has been developed. A particular HBP under study, known as HB560, has been characterized and successfully integrated with microelectromechanical systems processing. An array of electroplated nickel cantilever beams 100 μm×100 μm ~1000 μm in length has been successfully fabricated and dry released. The required release time is 10 min at 600°C in O₂ . Single open-ended microchannels were fabricated to determine the etch time and tunneling length achieved using HB560 as a sacrificial material. The time improvement for the sacrificial release is on the order of 40× compared to standard HF wet-etch processes. Etch lengths of 1 cm have been successfully demonstrated with a sacrificial layer aspect ratio of 1600. A photosensitive version of the dendritic material has also been synthesized and characterized to reduce the process steps     

Mechanical stability and adhesion of microstructures undercapillary forces. II. Experiments

For pt.I see ibid., vol.2, p. 33 (1993). Strong capillary forces are developed in the fabrication process of surface micromachined structures during the wet etch of sacrificial layers. The magnitude of these forces is in some cases sufficient to deform and pin these structures to the substrate resulting in device failure. The deflection, mechanical stability, and adhesion of thin micromechanical structure under capillary forces are examined. Microstructure adhesion is considered and experimental data for polycrystalline silicon microstructures are presented     

Mechanical stability and adhesion of microstructures undercapillary forces. I. Basic theory

Strong capillary forces are developed in the fabrication process of surface micromachined structures during the wet etch of sacrificial layers. The magnitude of these forces is in some cases sufficient to deform and pin these structures to the substrate resulting in device failure. The deflection, mechanical stability, and adhesion of thin micromechanical structure under capillary forces are examined. These phenomena are divided into two separate stages of mechanical collapse and adhesion to the underlying substrate. The basic theory of collapse is described. Approximate conditions are computed to prevent contact to the substrate     

Sacrificial wafer bonding for planarization after very deep etching

A new technique is presented that provides planarization after a very deep etching step in silicon. This offers the possibility for resist spinning and layer patterning as well as realization of bridges or cantilevers across deep holes or grooves. The sacrificial wafer bonding technique contains a wafer bond step followed by an etch back. Results of polymer bonding followed by dry etching and anodic bonding combined with KOH etching are discussed. The polymer bonding has been applied in a strain based membrane pressure sensor to pattern the strain gauges and to provide electrical connections across a deep corrugation in a thin silicon nitride membrane by metal bridges     

A fun and educative way for students to learn the nanofabrication through hands-on processing: Microletters

A 12-h hands-on processing protocol for a nanofabrication lab module is developed with the intention of teaching high school to college undergraduate students major fabrication techniques. The nanofabrication techniques employed are thin film deposition, direct-write lithography, conventional UV lithography with a mask aligner, etch, dicing, and characterization. The lab module protocol concludes with the student fabricating their own microletter silicon chip with a personalized message with letters and images through hands-on processing. For academic discussion, the light interference in thin film, deposition and etch rate, and etch selectivity are studied during the lab module.