Collaborative Research: A Diagnostic and Modeling Investigation of Pulsed PECVD

这个合作研究项目的目标是通过先进诊断和高性能计算机模拟的协同结合，对脉冲等离子体增强化学气相沉积（PECVD）有一个基本的了解。等离子体功率的低频调制对于脉冲PECVD和等离子体增强原子层沉积（PE-ALD）等工艺至关重要。这些先进的沉积技术赋予纳米级的厚度和成分控制，这是推进纳米技术的关键。特别是，脉冲PECVD正被应用于氧化物结构，作为高性能电介质，光学元件和扩散屏障的许多应用。然而，目前工艺开发完全是经验性的。时间分辨Langmuir探针，阻抗光谱和发射光谱将与最先进的建模相结合，以便为这些系统中控制等离子体化学的动力学提供基本见解。在脉冲PECVD中，中性物质（流体流动、输运现象、气体和表面化学）和活性物质（离子、电子、亚稳态）的作用是相当重要的。再加上该过程的瞬态性质，这一具有挑战性的问题将需要开发新的技术，以确保高保真的测量和设计可处理的计算模型。建模的挑战将通过变革性的方法来解决，以加速具有大量物种和化学反应的等离子体模拟的计算。共享内存，数据并行计算方法将用于解决等离子体模拟问题，这些问题通常具有中等大小的网格，但由于大量的物种和反应而产生的大自由度。用于解决脉冲PECVD所带来的复杂问题的系统方法将为PE-ALD和常规PECVD提供基本的见解。这项工作的更广泛影响还将包括培养两名博士候选人，并在独特的跨学科环境中参与具有重要技术意义的领域的本科研究人员。一名学生将专注于等离子体建模和模拟，而另一名学生将发展等离子体诊断和沉积方面的专业知识。这些学生和本科REU参与者将共同开发一个脉冲PECVD分析的综合平台。这些合作项目将参与针对德克萨斯大学奥斯汀分校（University of Texas Austin）和科罗拉多矿业学院（Colorado School of Mines）代表性不足群体的现有教师拓展项目，共同开发K-12教育模块。他们还将合作开发和改进一门关于等离子体处理的研究生选修课程。

CBET-0829043WoldenThe goal of this collaborative research project is to develop a fundamental understanding of pulsed plasma-enhanced chemical vapor deposition (PECVD) through a synergistic combination of advanced diagnostics and high-performance computer simulations. Low-frequency modulation of plasma power is critical to processes such as pulsed PECVD and plasma-enhanced atomic layer deposition (PE-ALD). These advanced deposition techniques impart nm-level control over thickness and composition that is critical to advancing nanotechnology. In particular, pulsed PECVD is being applied to oxide structures that have numerous applications as high-performance dielectrics, optical components, and diffusion barriers. However, process development is completely empirical at present. Time-resolved Langmuir probe, impedance spectroscopy, and emission spectroscopy will be coupled to state-of-the-art modeling in order to provide fundamental insights into the dynamics controlling the plasma chemistry in these systems. In pulsed PECVD the role of both neutrals (fluid flow, transport phenomena, gas- and surface chemistry) and activated species (ion, electrons, metastables) are of comparable importance. Coupled to the transient nature of the process, this challenging problem will require the development of novel techniques to both ensure high-fidelity measurements and to design tractable computational models. The modeling challenges will be addressed through transformative approaches to accelerate computations for plasma simulations with large number of species and chemical reactions. A shared-memory, data-parallel computing approach will be used to address plasma-simulation problems that typically have moderate-sized meshes but large degrees of freedom resulting from the large number of species and reactions. The systematic approach used to address this complex problem posed by pulsed PECVD will provide fundamental insight into both PE-ALD and conventional PECVD. The broader impacts of this work will also include the training of two PhD candidates and the engagement of undergraduate researchers in areas of great technological importance in a unique interdisciplinary environment. One student will focus on plasma modeling and simulation, while the second develops expertise in plasma diagnostics and deposition. These students, along with undergraduate REU participants, will work together in developing an integrated platform for the analysis of pulsed PECVD. The co-PIs will participate in established faculty outreach programs directed at underrepresented groups at the University of Texas Austin and Colorado School of Mines, working together to develop K-12 educational modules. They will also collaborate to develop and improve a graduate elective course on plasma processing.