We investigate Si epitaxy using 3D reactor scale computational fluid dynamics simulations coupled with surface chemistry models for the growth of pure silicon and phosphorus-doped silicon (Si:P) films. We focus on low temperature Si and Si:P processes using dichlorosilane (DCS) and phosphine. Based on existing DCS-based Si chemistry models for higher process temperatures, we developed a new kinetic chemistry model for low temperature Si epitaxy. To include doping, we developed an additional empirical model for Si:P epitaxy as there is not sufficient qualitative data on phosphine chemistry available for a kinetic chemistry model. This work provides Si and Si:P surface chemistry models, which allow reactor scale process simulations to get valuable process insights, enabling rational process optimization and supporting process transfer. Process optimization is demonstrated through process parameter variation with the main goal being the reduction of Si process variability by increasing within-wafer growth rate homogeneity.

中文翻译：

---

单晶圆反应器中低温硅外延工艺模拟的表面化学模型

我们使用 3D 反应器规模计算流体动力学模拟以及表面化学模型来研究硅外延，以生长纯硅和磷掺杂硅 (Si:P) 薄膜。我们专注于使用二氯硅烷 (DCS) 和磷化氢的低温 Si 和 Si:P 工艺。基于现有的基于 DCS 的更高工艺温度的硅化学模型，我们开发了一种新的低温硅外延动力学化学模型。为了包括掺杂，我们开发了一个额外的 Si:P 外延经验模型，因为没有足够的磷化氢化学定性数据可用于动力学化学模型。这项工作提供了 Si 和 Si:P 表面化学模型，使反应器规模的工艺模拟能够获得有价值的工艺见解，从而实现合理的工艺优化并支持工艺转移。工艺优化通过工艺参数变化来证明，主要目标是通过提高晶圆内生长速率的均匀性来减少硅工艺的可变性。