层压
材料科学
复合材料
互连
涂层
铜
制作
图层(电子)
冶金
计算机网络
计算机科学
医学
病理
替代医学
作者
David A. Hutt,Karen Williams,Paul Conway,Fuad Khoshnaw,Xiaoyun Cui,Deepa Bhatt
出处
期刊:Circuit World
[Emerald Publishing Limited]
日期:2007-02-13
卷期号:33 (1): 22-30
被引量:12
标识
DOI:10.1108/03056120710723689
摘要
Purpose To present the aims and preliminary findings of a research project to investigate the manufacture of multilayer glass substrates built up from thin glass sheets. Design/methodology/approach The approaches that may be taken to create glass substrates and the challenges involved are described. Excimer laser machining was used for the formation of microvias and other features in individual glass sheets. In addition, methods for the electroless copper metallisation of the smooth glass surfaces were studied. Finally, a technique for the lamination of the glass layers using low temperature, pressure assisted bonding was investigated. Findings Microvias with 100 μ m diameter entry holes were successfully machined in 100 μ m thick glass sheets and process windows were identified to reduce debris and hole taper. Using appropriate pre‐treatment steps, electroless copper coatings could be deposited uniformly over the smooth glass surface, however, further improvements in adhesion were found to be necessary. The direct lamination of glass layers was found to be possible using pressure and temperature applied over long periods of time. Improvements to the lamination process were made to reduce the initiation of cracks which were assessed using fatigue testing. Research limitations/implications The feasibility of the individual steps in the fabrication of glass substrates has been demonstrated. Further work is necessary to control the processes in order to limit microcrack formation, improve copper coating adhesion and ensure uniform lamination of multiple glass layers. Originality/value The use of glass materials could enable the manufacture of substrates for high density electrical interconnect with integrated optical waveguides.
科研通智能强力驱动
Strongly Powered by AbleSci AI