Etching of Dielectrics: Fundamental Plasma-Surface Interactions Through Mass-Filtered, Energy-Tuned Ion Beams

本研究旨在了解在氟碳等离子体或其他适当的化学物质中蚀刻硅、二氧化硅（二氧化硅）和其他新型介电材料（氧化锆、多孔氧化硅、黑金刚石）的基本等离子体表面相互作用。为此，利用结合散射和表面诊断的独特离子束线，等离子体提取、质量过滤的可调能量离子束被定向到上述介质材料的表面。相同或不同化学成分的中性自由基也以不同的中性离子比同时供应。发射的反应产物由飞行时间四极杆质谱法实时监测。利用表面分析技术对相互作用的稳定表面产物进行了分析。与硅表面的散射进行比较，以了解介电材料在硅上的选择性蚀刻。具体目标包括了解与半导体工业相关的表面上的高能、惰性和活性离子的散射动力学；氟碳等离子体中硅与硅之间蚀刻选择性的原因区分非热反应途径并测量它们相对于热反应对蚀刻的贡献；监测离子轰击促进的表面修饰的程度和化学性质，作为离子的平移能和化学特性的函数；测量蚀刻产率随抛射离子的平动能和入射角的函数，并建立观察到的依赖关系的现象学模型；利用光束散射的腐蚀屈服模型模拟剖面演化；并在光束路径上进行蚀刻实验，以进行比较和验证。受控离子束表面实验的目标是在高密度等离子体中蚀刻硅、二氧化硅和新型介电材料时发生的基本相互作用。产生了散射相互作用的详细图像，包括蚀刻产率和反应产物作为入射能量和角度的函数。复合离子[CFx+和SiFx+ (x=1-3)]是用于蚀刻氧化物和低k电介质的典型复合氟碳化学物质。快速筛选了一些离子在蚀刻各种材料中的作用。这建立了一个新的范例，选择等离子体化学来实现所需的蚀刻速率和蚀刻轮廓，而不是现在工业实践的耗时和昂贵的试错配方开发。这些实验被用来验证分子动力学模拟的结果，以提高人们的理解，并产生行业所需的散射现象学模型，以真正预测剖面的演变。更广泛的影响所产生的结果是基本的，足以被理论家用于验证光束表面相互作用的模拟，并且足够实用，可用于工艺工程师在选择化学物质和操作条件时，允许快速优化蚀刻工具。通过基本光束散射实验和蚀刻轮廓演变模拟相结合，建立了蚀刻工艺发展的新范式。所获得的知识和理解被纳入等离子体表面相互作用的课程和教程，以教育学生和工程师化学反应动力学的基本原理。一个新的，低成本的实验涉及大气微等离子体在硅的直接图像化开发，以介绍等离子体表面相互作用的本科生。

SummaryThis research seeks to understand the fundamental plasma-surface interactions responsible for the etching of silicon, silicon dioxide (silica), and other novel dielectric materials (zirconia, porous silicon oxide, black diamond) in fluorocarbon plasmas or other appropriate chemistries. To this end, plasma-extracted, mass-filtered ion beams of adjustable energy are directed at surfaces of the aforementioned dielectric materials using a unique ion beamline combined with scattering and surface diagnostics. Neutral radicals of the same or different chemical composition are also supplied simultaneously at varying neutral to ion ratios. The emitted reaction products are monitored by time-of-flight quadrupole mass spectrometry in situ and in real time. The stable surface products of the interaction are analyzed using surface analysis techniques. Comparisons are made with scattering at silicon surfaces to understand selective etching of dielectric materials over silicon. Specific goals include the understanding of the scattering dynamics of energetic, inert, and reactive ions on surfaces of relevance to the semiconductor industry; the reasons for etch selectivity between silicon and silica in fluorocarbon plasmas; distinguishing non-thermal reaction pathways and measuring their contribution to etching relative to that of thermal reactions; monitoring the extent and chemical nature of the surface modification facilitated by the ion bombardment as a function of the translational energy and chemical identity of the ions; measurement of etch yields as functions of translational energy and incident angle of the projectile ions and developing phenomenological models of the observed dependencies; using the etch-yield models for beam scattering to simulate profile evolution; and performing etch experiments in the beam path for comparison and validation purposes.Controlled ion-beam surface experiments are performed that target the fundamental interactions occurring when etching silicon, silicon dioxide, and novel dielectric materials in high-density plasmas. A detailed picture of the scattering interaction is produced, including etch yields and reaction products as functions of incident energy and angle. Complex ions [CFx+ and SiFx+ (x=1-3)] typical of the complex fluorocarbon chemistries employed in the etching of oxides and low-k dielectrics are employed. A rapid screening of the role of a number of ions in etching various materials is performed. This establishes a new paradigm to select the plasma chemistry to achieve a desired etch rate and etch profile in contrast to the time-consuming and costly recipe development by trial-and-error now practiced by industry. The experiments are used to validate results from molecular dynamics simulations for improved understanding and to produce phenomenological models of scattering needed by the industry for truly predictive profile evolution.Broader impactThe results produced are fundamental enough to be used by theorists in the validation of simulations of beam-surface interactions and practical enough to be useful to process engineers in the selection of chemistries and operating conditions that permit rapid optimization of etch tools. A new paradigm in etch process development is established through a combination of fundamental beam-scattering experiments and etch-profile-evolution simulations. The knowledge and understanding obtained is incorporated in courses and tutorials on plasma-surface interactions to educate students and engineers on the underlying principles of chemical reaction dynamics. A new, low-cost experiment involving atmospheric microplasmas in direct patterning of silicon is developed to introduce plasma-surface interactions to undergraduates.