摘要
硅通孔(TSV)三维封装因其独特的工艺而备受关注,然而内部缺陷的检测一直是限制其进一步发展的难题。主动红外热成像技术是一种新型无损检测方法,具有无接触、高效率等优点,为实现对TSV内部典型缺陷的识别与分类,提出了一种基于激光加热主动激励的TSV内部缺陷分类识别方法。以激光为辐射热源,充分激发TSV内部缺陷,通过理论与仿真分析,掌握不同内部缺陷在主动激励下的外部温度分布表现规律;建立卷积神经网络模型,通过对外部温度分布结果的训练,实现内部缺陷的分类识别。通过试验证明,该方法对典型TSV内部缺陷具有良好的识别能力,识别准确率可达97.12%。利用主动红外热成像检测方法实现了对TSV内部缺陷的有效检测,为三维封装缺陷检测提供了一种快速有效的方法。
Through-silicon-via(TSV)3D packaging has been attracted much attention for its unique process technology.However,detection of internal defects has always been a challenge that hinders its further development.Active infrared thermography is a new non-destructive testing method with the advantages of non-contact and high efficiency.Here a method was proposed to identify and classify the typical internal defects in TSVs,which is based on laser-induced active stimulation.The laser was used as a radiative heat source to fully excite the internal defects within the TSVs.Through theoretical and simulation analysis,the performance characteristics were identified by studying the external temperature distribution for different internal defects under active stimulation.A convolutional neural network model was established and trained on the external temperature distribution to achieve the purpose of classification and recognition of internal defects.The experimental results demonstrate that the method is very effective to recognize typical internal defects in TSV,with an accuracy of 97.12%.This effective detection,using active infrared thermography,provides a rapid and efficient method for 3D packaging defect inspection.
作者
聂磊
刘江林
于晨睿
骆仁星
张鸣
NIE Lei;LIU Jiangin;YU Chenrui;LUO Renxing;ZHANG Ming(School of Mechanical Engineering,Hubei University of Technology,Wuhan430068,China;Hubei Taihe Electric Co.,Ltd.,Xiangyang441057,Hubei Province,China)
出处
《电子元件与材料》
CAS
北大核心
2023年第12期1521-1528,共8页
Electronic Components And Materials
基金
国家自然科学基金(51975191)。
