Plasma based process control of reactive sputtering

反应溅射是在低温等离子体中沉积薄膜的常用方法，是各种高科技产品的基础。尽管它有巨大的社会影响，但许多基本的机制，如电子加热、等离子体-表面相互作用和输运现象，在这种等离子体中并不清楚。因此，还缺乏对中性气体压力、气体流量、功率、磁场等全局控制参数对沉积薄膜特性的非线性和非局部影响的深入研究。这导致了过程优化和控制的巨大局限性。由于缺乏对决定膜特性的不同颗粒种类的能量分布函数以及表面特性本身的理解和访问，当前的工艺控制概念基于控制诸如气体辉光和功率的辅助参数，其不能唯一地确定膜特性，以及灰箱模型。在这个意义上，等离子体溅射工艺是存在相同问题的各种其他等离子体工艺的典型例子。在本项目中，将首次开发和测试一种基于等离子体的概念，即基于等离子体参数和科学认识来控制反应溅射以在电容射频(RF)磁控等离子体中沉积Al_2O_3薄膜。基于实验和理论方法的协同结合，我们将对全局控制参数对不同粒子种类的能量分布函数的影响以及膜特性本身的影响有一个基本的理解。放电将通过各种等离子体和表面诊断以及动力学数值模拟来实验表征。在此基础上，将确定适用于过程控制的有能力的等离子体参数。这些参数必须可以在几秒钟内通过非侵入性诊断进行测量，并且必须与能量分布和表面特征唯一相关。它们将通过快速简化的全球模型在理论上进行描述，这些模型可以预测寄生过程漂移的原因和补偿它们的程序。这些等离子体参数和模型将在并行开发的控制回路框架内用于过程控制。这种基于等离子体的控制回路将显著改善对所沉积薄膜的特性的控制。目标是控制等离子体特性，如每个沉积粒子的能量，以使薄膜和等离子体特性相互关联，并最终控制薄膜特性。如果成功，这种基于等离子体的过程控制原理可以转移到其他等离子体过程中。此外，对这类等离子体的物理的重要的基本见解也是可以期待的。

Reactive sputtering is frequently used to deposit thin films in low temperature plasmas as the basis for a variety of high-tech products. Despite its enormous societal impact many fundamental mechanisms such as the electron heating, plasma-surface interactions, and transport phenomena are not understood in such plasmas. Consequently, there is also a lack of insights into the non-linear and non-local effects of global control parameters such as the neutral gas pressure, gas flows, powers, magnetic fields, etc. on the characteristics of deposited thin films. This leads to massive limitations of process optimization and control. Due to a lack of understanding of and access to the energy distribution functions of different particle species, that determine the film properties, as well as to the surface characteristics themselves current concepts of process control are based on controlling secondary parameters such as gas glows and powers, which do not determine the film characteristics uniquely, as well as greybox-models. In this sense plasma sputter processes are a representative example of a variety of other plasma processes that suffer from the same problems.In this project, a plasma based concept, i.e. based on plasma parameters and scientific understanding, to control reactive sputtering to deposit Al2O3-films in a capacitive radio frequency (RF) magnetron plasma will be developed and tested for the first time. A fundamental understanding of the effects of global control parameters on energy distribution functions of different particle species, that determine the film characteristics, as well as of the film characteristics themselves will be developed based on a synergistic combination of experimental and theoretical methods. The discharge will be characterized experimentally by a variety of plasma and surface diagnostics as well as numerically by kinetic simulations. Based on this fundamental understanding capable plasma parameters suitable for process control will be identified. These parameters must be measurable within seconds by non-invasive diagnostics and must be uniquely correlated to energy distributions and surface characteristics. They will be described theoretically by fast reduced global models that can predict causes of parasitic process drifts and procedures to compensate them. These plasma parameters and models will be used for process control in the frame of a control loop that is developed in parallel. This plasma based control loop will result in a significant improvement of the control of the characteristics of deposited thin films. The goal is to control plasma characteristics such as the energy per deposited particle to correlate film and plasma characteristics and to ultimately control the film characteristics. If successful, this principle of plasma based process control can be transferred to other plasma processes. Moreover, important fundamental insights into the physics of such plasmas are expected.