Ultra-High-Resolution Neural Sensing Biosensor Development with 3D Heterogeneous Integration

• Main Authors: Hu, Yu-Chen, 胡毓宸
• Other Authors: Chen, Kuan-Neng
• Format: Others
• Language: en_US
• Published: 2017
• Online Access: http://ndltd.ncl.edu.tw/handle/cxg5z7

博士 === 國立交通大學 === 電子研究所 === 105 === In this thesis, 3D IC technology is adopted to the neural sensing microstructure. Three different biosensors are proposed and demonstrated with different packaging approaches and animal experiment applications. 3D-SiP neural sensing biosensor, 2.5D-silicon interposer neural sensing biosensor, and 2.5D-flxeible interposer neural sensing biosensor are investigated in this thesis. With the highly integrated neural sensing microstructure, the three novel biosensors are expected to contribute to the biomedical field through identifying and solving unknown biological mysteries. In recent years, neuroscience has been a research area that is valued by many countries. One of the key technologies of that area that is widely interested in is the approach on an effective way of extracting neural signals. Therefore, many neural sensing microstructures have been proposed with the aim of fabricating a well-functional biosensor. A powerful functional neural sensing microstructure is critical in order to detect neural signals with low noise and high spatial resolution. Besides, the miniaturized packaging is also necessary since the mouse experimented on moves around freely during the experiment. In this research, the ultra-high-resolution demand is satisfied with the silicon semiconductor standard process, and the miniaturized packaging is achieved with the 3D heterogeneous integration technology. These biosensors can be divided into three main parts: (1) TSV-embedded electrode probe, (2) interposer, (3) neural sensing circuits. In accordance to the difference of functional aspects, three different packaging biosensors are fabricated in this thesis. TypeⅠ: 3D-SiP neural sensing biosensor Two circuit chips are bonded face-to-face while wire bonding is used to connect the bottom chip to an interposer. The TSV-embedded electrode probe transmits neural signals through TSV and RDL on the interposer to the connected wires. With the advantages of low cost and short development time, SIP is the first packaging approach to be implemented in the neural sensing microstructure. In this development phase, chip-level heterogeneous integration platform is also developed to overcome the over-expensive issue in 3D wafer-level packaging. TypeⅡ: 2.5D-silicon interposer neural sensing biosensor Silicon interposer is the key component of this type. Interposer carries multiple neural sensing circuit chips and does the routing paths between the front side chips and backside TSV-embedded electrode probe with TSV and RDL. As compared to type Ⅰ, flip-chip approach is adopted in this type to further miniaturized the packaging size. In this development phase, biocompatible Au-TSV is developed as the measuring tool that is safer for the living organisms. Type Ⅲ: 2.5D-flxeible interposer neural sensing biosensor In order to fulfill the demand of animal experiment, silicon interposer is replaced with polyimide for better flexibility and lighter. In addition, a novel silicon-polyimide bonding approach is developed for better electrical property than traditional anisotropic conductive film or non-conductive paste bonding approaches. All the biosensors are tested with electrical examinations, including Kelvin structure, Daisy chain, and Comb structure. Mechanical strength test is adopted to ensure the robustness of the structure. Reliability tests are evaluated including thermal cycling test (TCT), un-bias highly accelerated stress test (un-bias HAST) and multiple current stressing test. Material analysis and process developments are detailed depict in this thesis. Overall, 3D IC technology is introduced into biosensor development in this research for ultra-high-resolution. The novel heterogeneous integration structure is a breakthrough in both the medical field and electronic field, and will contribute to multiple disciplines in the future.