Long,
What you are experiencing is a phenomenon called gold spitting.  It is
common when evaporating gold to see this.  In fact, if your evaporator has a
window that looks toward the e-gun you will see these spits ejected from the
melt.  Only the smallest ones get to the wafer,  you will probably find
"larger" ones in your chamber.  You can try a couple of things to solve it:
try a lower deposition rate; or, if you use a beam sweep, make sure that it
covers a large area on your melt.  If that fails, you can look for a paper
written in the 1970s that suggests adding a small piece of tantalum to your
gold.  The Ta will not melt but it will control the spitting.  I don't
remember the author's name, but I am pretty sure it was someone at IBM.
Brad


On 11/3/05 9:30 AM, "Long Chen"  wrote:

> Hi, I am doing some gold evaporation using e-beam evaporator, and
> consistently I am getting defects in the resulted film. In the SEM the
> defects look like pretty round balls with diameters about 100~300nm.
> After putting in the ultrasonic bath for a few seconds, those defects
> would partially move out and I could see that the holes beneath. I
> didn't have the tool to check the element of the defects but found a
> technique paper from Motorola which discussed about the similar defects
> happened in their gold evaporation. It said that the defects are "spit"
> gold where the carbon contaminants on the target surface act as the
> spitting catalyst. They tried several approaches to reduce the carbon
> amount in the target material and the density of the defects decreased
> accordingly. I am wondering whether some guys here know about this and
> have some idea to deal with this, i.e., to improve the film quality. Any
> suggestion would help and would be deeply appreciated.