Here is the abstract you requested from the socal_2015 technical program page. This is the original abstract submitted by the author. Any changes to the technical content of the final manuscript published by IMAPS or the presentation that is given during the event is done by the author, not IMAPS.
|Diffusion Controlled Intermetallic Growth in Cu-Al Wire Bonds|
|Keywords: intermetallic, Cu-Al, Wire bond|
|Mechanism of Intermetallic Compounds (IMCs) formation during wire bonding processes, and their post-bonding growth play an important role in reliability of integrated circuits. Since bond formation occurs over a very short period of time, success of pure experimental analysis in identifying micromechanisms of IMC formation is rather limited. To address the drawback of experimental approaches, in this research a combination of finite element computational analysis and analytical investigation is used. Finite Element Analysis (FEA) is used to simulate stress distribution in the capillary and aluminum pad. A coupled thermal-mechanical finite element stress analysis is developed for ball bonding of copper wire to aluminum pad. Stress distribution in the wire and aluminum pad is calculated adapting a “non-linear large deformation analysis” in ANSYS. Using analytical approaches, the stress calculated through FEA is related to dislocation density that is generated in the aluminum pad during the wire bonding process. Dislocation density is incorporated in diffusional equations. Subsequently, diffusion rate in the presence of desolations is computed. Adapting two main assumptions: 1) cupper being the main diffusant element and 2) Al2Cu being the first intermetallic compound that forms between the copper wire and aluminum pad, size of Al2Cu particles immediately after wire bonding is calculated. The same approach is used to predict intermetallic growth at elevated service temperatures. The results are compared with existing information in the literature. Validity of the developed approach is confirmed by the good agreement between the predicted results and the independently acquired experimental data.|
Santa Clara University
Santa Clara, CA