Photomask Defect Inspection and Metrology for Semiconductor Lithography Technology

半导体是智能手机和电器等众多产品不可或缺的组成部分，该奖项有助于为半导体行业创造光掩膜检测和计量方面的新知识。干涉检测技术的出现有助于提高光刻制造分辨率，促进微型化，提高电子器件的性能和降低成本。在半导体制造工艺中，带有孔和透明度的光罩允许光以特定的模式照射，这是光刻技术的一种使能技术。光掩模缺陷，如不需要的图案、污染和衬底缺陷，成为光刻工艺和图案不准确的日益重要的来源。该奖项支持在光掩膜检测、计量、设计和制造方面开发新的测量原理的基础研究。该项目的检测技术和模拟方法能够确定掩膜质量。该方法识别光掩模缺陷，建立缺陷模型并补偿光刻制造误差。因此，这项研究通过自动化光掩膜缺陷分析过程，为光掩膜制造商节省了大量的时间和成本，这有利于光掩膜、光刻机床和半导体制造行业，从而有利于美国经济。该项目为学生提供跨学科的研究经验，通过精密计量实习项目扩大代表性不足群体的研究参与，并通过机械测量和精密机床课程影响工程教育。随着光掩膜行业向越来越小的技术节点发展，对更积极的检测的需求变得至关重要。光掩膜检查包括检查用于半导体器件制造的光掩膜的正确性。有缺陷的光掩膜的投影图像表现为不规则，例如，线宽度比设计的更窄或更宽，图案的线边缘粗糙度以及可能显著改变正在制造的机械和电气性能的扭曲。这个问题变得更加明显，因为光掩模的复杂性随着新半导体芯片的复杂性而上升。本研究利用刀口衍射干涉法（KEDI）和人工神经网络模型填补了掩膜缺陷检测技术和基础识别的关键知识空白。通过从严格的模拟工具和实验结果中获得的深刻认识，研究团队确定光刻光掩膜缺陷的可印刷性可以在光刻操作之前定量表征，并且可以扩展模拟模型，以补偿使用小于光波长的印刷结构在亚波长光刻期间发生的图像畸变。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。

Semiconductors are the indispensable components of a wide range of products, such as smartphones and electrical appliances, and this award contributes to the creation of new knowledge in photomask inspection and metrology for the semiconductor industry. The availability of the interferometry inspection technology helps enhance the lithography manufacturing resolution, promote miniaturization and increase the performance and reduce the cost of electronic devices. In semiconductor manufacturing processes, photomasks with holes and transparencies that allow light to shine through in a defined pattern are an enabling technology for lithography. Photomask defects, such as unwanted patterns, contaminations and substrate flaws, become increasingly important sources of inaccuracies in the lithography processes and patterning. This award supports fundamental research to develop new measurement principles in photomask inspection, metrology, design and manufacturing. The project's inspection technique and simulation methods enable the determination of photomask quality. The method identifies photomask defects, creates defect models and compensates for lithography manufacturing errors. As a result, this research provides photomask manufacturers with significant time and cost savings by automating the photomask defect analysis process, which benefits the photomask, lithography machine tool and semiconductor manufacturing industries and, thus, the U.S. economy. This project provides interdisciplinary research experience for students, broadens participation of underrepresented groups in research through a precision metrology internship program, and influences engineering education through a course in Mechanical Measurements and Precision Machine Tools. As the photomask industry progresses towards smaller and smaller technology nodes, the need for more aggressive inspection becomes critical. Photomask inspection involves checking the correctness of the fabricated photomasks used for semiconductor device fabrication. The projected images of the defective photomask appear as irregularities, such as, line widths that are narrower or wider than designed, line-edge roughness of the patterns and distortions that may significantly alter the mechanical and electrical properties of what is being fabricated. This problem becomes more pronounced since photomask complexity rises with the complexity of new semiconductor chips. This research fills a critical knowledge gap in inspection technology and fundamental identification of photomask defects by using knife-edge diffraction interferometry (KEDI) and artificial neural network models. Through insights obtained from rigorous simulation tools and experimental results, the research team determines that the printability of lithography photomask defects can be quantitatively characterized prior to lithography operation and the simulation model can be extended to compensate for image distortions that occur during sub-wavelength lithography using printed structures smaller than the wavelength of light.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.