Plating Into Solution

I have a question concerning what plating will go into solution during the reflow process. I am soldering a small piece of ceramic (120x180 mils 7 mil thick) onto a flex circuit. I am seeing some fractures that I am sure are being caused by handling. Some unfortunately appear and I just can't see a handling issue as the cause. The ceramic is quite thin and is fragile and does fracture if mishandled which is the failure mode we see normally, it is the ones where the ceramic does not break....it just separates from the flex that is the issue. I am using Indium 5.1AT and a pretty normal ramp to spike profile. The ceramic is from EDO Ceramics and is EC-65 Lead Zirconium Titanate Piezoelectric. We are slicing it with high precision dicing saws. Then we sputter the following: 1) Titanium 600 to 1200 Angstroms 2) Copper 8500 to 11,000 Angstroms 3) Nickel 1600 to 2100 Angstroms 4) Gold 1800 to 2100 Angstroms I know that the gold and nickel are going into solution during the reflow process. I don’t believe that the copper is only because of its thickness. If for some reason I am pulling all of the copper off what would be the intermetallic created with titanium? Is is a strong bond or a very weak bond? I would appreciate any thoughts on this.

reply »