Abstract Preview

Here is the abstract you requested from the dpc_2018 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.

Thermosonic Ball Bonding Recipe Optimization: Comparing Cu and PCC Wire on Two Pad Thicknesses
Keywords: Copper wire bonding, Process optimization, Bond geometry
Cu wire bonding results in cost savings in microelectronics packaging, including automotive packaging. Improved wires, equipment, and procedures have helped to overcome many of the process-related challenges of Cu wire bonding, including bondability, small process windows, and reliability [1]. However, the cost savings due to the Cu material is partially offset by the high level of process control required in production. More robust bonding processes would result in more relaxed manufacturing conditions or higher yield and quality. Another challenge is to find ways to use Cu wire bonds for high operating temperatures in automotive applications, e.g. 200 C. To address these challenges, it would be helpful to improve the knowledge of physical mechanisms responsible for Cu bond formation and subsequent degradation in accelerated testing and operation. In the past years, much progress was made in understanding Cu wire bonding better. Studies in this direction have included detailed analysis of interfacial features and their evolution during accelerated testing. Effects of various bonding process factors and wire material coatings (e.g. Pd) have been described. However, results depend on the size of the bonds. For example, large bonds behave differently in reliability test compared to smaller bonds. Studies generally do not standardize results to the bond size, and comparisons including different bonded ball sizes are very rare. For a comprehensive understanding of Cu bond formation and reliability, the bonded ball size must be included as a determining factor, and methods need to be applied to control the size when studying the effects of various parameters. In this study, bond size is controlled using a bonding recipe optimization method developed previously at the Centre for Advanced Materials Joining in the University of Waterloo [2]. With two different control targets for the bond size, its interaction effects with other factors is studied, including wire type and pad thickness. Two types of 25 m diameter wire are thermosonically ball bonded on industry-provided aluminum-copper (Al- 0.5%Cu) pads on silicon test die. Wire materials are bare Cu or palladium-coated copper (PCC). Bond ball target diameters are 61 m and 75 m for both wire types, with 14 m and 17 m ball heights respectively, using the same capillary. Bond recipe optimizations were accomplished on 3 m pad Al thickness. Wire bonds were then created using the optimized recipes on both 3 m and 0.8 m thick pad Al for analysis. Measurements include shear test and physical and dimensional analyses from bond cross sections. The bonding recipe optimizations involve tuning two bond recipe parameters, electronic flame- off (EFO) current and impact force settings to obtain approximately the desired ball geometry at 175 C temperature and a fixed time parameter, then maximizing shear force while keeping Al splash and ball geometry under control by adjusting the ultrasonic power. Data from these experiments facilitate future work in assessing the reliability of these ball bonds.
Michael D. Hook, Ph. D. Candidate
University of Waterloo
Waterloo, ON

  • Amkor
  • ASE
  • Canon
  • Corning
  • EMD Performance Materials
  • Honeywell
  • Indium
  • Kester
  • Kyocera America
  • Master Bond
  • Micro Systems Technologies
  • MRSI
  • Palomar
  • Promex
  • Qualcomm
  • Quik-Pak
  • Raytheon
  • Rochester Electronics
  • Specialty Coating Systems
  • Spectrum Semiconductor Materials
  • Technic