SBIR Phase II: Disruptive Semiconductor Software Tool for Recipe Optimization for Deposition and Etching Processes

小型企业创新研究(SBIR)第二阶段项目的更广泛影响/商业潜力是实现半导体设备的下一代制造技术。新型半导体器件的部署经常受到纳米制造工艺的发展的阻碍，例如为闪存开发的最新器件架构3D NAND。这项提议的目标是建立在配方预测平台的基础上，为半导体设备和芯片制造商创建独立的软件解决方案，以快速优化其工艺。在这项研究中，将提高软件速度，将预测能力扩展到3D结构，并将模型校准自动化，以方便客户部署。通过加快器件制造工艺的开发，该项目降低了半导体行业配方开发的成本，并实现了下一代技术。该小型企业创新研究(SBIR)第二阶段项目将开发新的高级物理和统计模型，以快速准确地预测等离子刻蚀和沉积等工艺的结果。在这项研究中，2D和3D轮廓模拟将在算法上进行速度优化，为工艺工程师提供即时结果。机器学习模型将用于使流程工程师能够快速探索复杂的贸易空间。将开发复杂的数值算法，以帮助工程师最大化工艺窗口。基于统计和物理的工具的创新将允许过程工程师自己构建和校准模型，从而支持下一代技术。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase II project is the enablement of next generation manufacturing technologies for semiconductor devices. The deployment of novel semiconductor devices such as 3D NAND, the latest device architecture developed for flash memory, is often stymied by the development of nanomanufacturing processes. The objective of this proposal is to build upon a recipe prediction platform to create a standalone software solution for semiconductor equipment and chip manufacturers to rapidly optimize their processes. In this research, the software speed will be increased, prediction capabilities will be extended to 3D structures, and model calibration will be automated to facilitate customer deployment. By accelerating the development of the processes used in device fabrication, this project reduces the cost of recipe development in the semiconductor industry and bringing next-generation technologies to fruition.This Small Business Innovation Research (SBIR) Phase II project will develop new advanced physical and statistical models to quickly and accurately predict outcomes for processes like plasma etch and deposition. In this research, 2D and 3D profile simulations will be algorithmically optimized for speed to give process engineers instantaneous results. Machine learning models will be used to enable process engineers to rapidly explore complex trade spaces. Sophisticated numerical algorithms will be developed to help engineers maximize process windows. Innovations in statistical and physics-based tools will permit process engineers to build and calibrate models themselves, enabling next-generation technologies.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.