CAREER: Continuous Flow Chemistry of Microelectronics Polymers via Combined Physics-based and Machine Learning Models

微电子聚合物在半导体技术的发展中是必不可少的。精确的微芯片制造依赖于高质量的聚合物输入。然而，由于放大、工艺控制、纯度、质量控制和生产时间等方面的挑战，目前生产微电子聚合物的方法并不能满足高通量制造的所有要求。为了应对这些挑战，PI计划创建一个平台，用于数据驱动的聚合物材料及其制造工艺设计，以及这些工艺的最佳操作和控制。PI的独特方法是用连续流动反应器取代目前大容量间歇式反应器的做法，从而可以精确控制聚合物的性质和结构。提议的平台是环保的，因为它承诺减少碳足迹，降低运营成本，并限制在制造过程中产生的劣质、不可用材料的数量。该项目还将推进其他几个领域的知识，因为从该项目中获得的建模和制造知识将适用于其他特种聚合物。除了培训研究生和本科生进行研究外，该项目还将促进先进制造业课程材料的开发，并将涉及以先进制造业和微电子技术为重点的高中教师培训形式的推广活动。该项目将解决与微电子聚合物制造相关的几个基础研究问题。研究人员将研究反应条件、工艺参数和原材料质量对微电子聚合物性能和质量属性的影响。将进行实验和理论/计算研究，以实现微电子聚合物的连续流动化学。将开发基于物理和机器学习的过程模型，以预测生产这些聚合物的过程的动力学。这些模型将考虑热量、质量、动量传递和反应动力学，以预测过程变量，如聚合物分子量和聚合物组成。这些模型也将有助于更好地理解流动反应器中合成的微电子聚合物的反应机理和结构-性能关系。因此，预计该模型将减少化学维度空间，以加快发现新的微电子聚合物，满足所需的分子结构-性质关系。该项目的一个可交付成果是一个制造平台，该平台使用来自反应监测和操作光谱的在线信息来反馈控制聚合物质量属性，从而实现微电子聚合物的高效连续和自主生产。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Microelectronics polymers are essential in the development of semiconductor technology. Precise microchip manufacturing hinges on high-quality polymer inputs. However, current methods for producing microelectronics polymers do not satisfy all the requirements of high throughput manufacturing due to challenges with scale-up, process control, purity, quality control, and production time. To tackle some of these challenges, the PI plans to create a platform for data-driven design of polymer materials and their manufacturing processes, as well as optimal operation and control of these processes. The PI’s unique approach is to replace the current practice of high-volume batch reactors with continuous flow reactors that allow for the precise control of polymer properties and structure. The proposed platform is eco-friendly, as it promises to reduce the carbon footprint, decrease operating costs, and limit the amount of inferior, unusable materials generated during the manufacturing. The program will also advance knowledge across several other fields, as the modeling and manufacturing knowledge gained from this project will be applicable to other specialty polymers. In addition to training graduate and undergraduate students in research, the project will contribute to the development of course materials on advanced manufacturing and will involve outreach activities in the form of high school teacher trainings focused on advanced manufacturing and microelectronics.This project will address several fundamental research problems related to microelectronics polymers manufacturing. Researchers will investigate the effect of reaction conditions, process parameters, and quality of raw materials on the microelectronics polymers properties, and quality attributes. Experimental and theoretical/computational studies will be performed to enable the continuous flow chemistry of microelectronics polymers. Combined physics-based and machine learning process models will be developed to predict the dynamics of the processes that produce these polymers. These models will account for heat, mass, and momentum transfer and reaction kinetics to predict process variables such as polymer molecular weight and polymer compositions. These models will also provide better understanding of reaction mechanisms and structure-property relationship of microelectronics polymers synthesized in flow reactors. Therefore, it is expected that the models will reduce the chemical dimensional space to expedite the discovery of new microelectronics polymers that meet the required molecular structure-property relationships. A deliverable of this project is a manufacturing platform that uses online information from reaction monitoring and in operando spectroscopy for feedback control of polymer quality attributes, allowing for the efficient continuous and autonomous production of microelectronics polymers.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.