| Summary: | 碩士 === 國立交通大學 === 管理學院工業工程與管理學程 === 107 === The development of science and technology has always come from human nature. The requirements of the most end-users for electronic products are no longer only required to be miniaturized, light and thin, but also high speed and diversified functions. The technology of the gold bump process on the driver IC has been developed for more than 20 years, and the related technology is quite mature. The design of high I/O pin count and shrinking die pad size and pitch has become a trend of IC development. Compared with the gold bumps, the pitch and spacing are continuously reduced, and the number of rows for the multi-function gold bump design has evolved from two rows to four rows. In the gold bump process production process of the four rows of gold bumps, the most difficult factor is the parameter setting value of the photoresist adhesion strength (magnification characteristic) in the photoresist process. The greater the photoresist adhesion strength, the more It is attacked by chemical agents in the post-process, and if the photoresist adhesion strength is insufficient, it will be attacked by chemical in the process, resulting in an increase in gold bump defects (under plating), and the shipment yield does not meet customer requirements, thus affecting the delivery date. Therefore, how to strengthen the adhesion in the photoresist process is an important topic for the gold bump industry. Therefore, the main purpose of this study is to find out the best parameter setting values that can maximize the adhesion of photoresist by using the current process equipment for the production of four rows of gold bump products through Design of Experiments (DOE). In order to reduce the gold bump defects (under plating) caused by insufficient adhesion strength of the photoresist for reaches the customer's shipping request.
Keyword: Gold bump process, Coating process, Design of Experiments(DOE),
Response surface methodology(RSM) ,Box-Behnken design (BBD)
|
|---|
