Where the
world gathers for metal finishing
Q&As since 1989


-----

Nickel diffusion through sputtered and plated gold

2004

We are observing a stain on a sputtered Au surface following one of our process steps. The metallization is 300nm Ti, 600nm NiV (< 10%wt V), 300nm Au (surface), all sputtered in situ. The stain was analyzed by both Auger ES and EDS to be Ni. The processing step involves soaking the wafers in an organic solvent heated to 100 degrees C for 12 hours. I wish I knew more about the chemistry of the organic solvent, but it is proprietary from our vendor. There is lots of topography on the wafer at this point in our processing, so the step coverage of the Ti/NiV/Au metal stack is not continuous, so we haven't ruled out some chemical interaction with the Ni that is then precipitating onto the surface of the Au. But I want to rule out Ni diffusion through the Au to its surface, if possible. We have electrolytically plated Au over the sputtered Au, and the problem (Ni on Au surface) is eliminated, but I don't know if it's due to sealing the metal so that none of the organic solvent makes contact with NiV or if it has to do with Ni diffusion differences between sputtered and plated Au. Any help with sputtered vs. plated Au (gold sulfite bath) grain size differences and Ni diffusion would be a great help.

Sincerely,

Terry Daly
Freescale Semiconductor - Tempe, Arizona, US

---

Could you talk about the electrical conditions in sputtering (DC, Pulsed, Rf) and whether there is a substrate bias?

John Carlotto
E. Providence, Rhode Island
2004