Hi,

For years we sputtered a thin layer (~250A) of TiW on the backside of some
wafers (mostly N type Phosphorous or P type Boron doped).
These wafers were later mounted on sticky tape, diced and then had their
chips picked and placed on waffle trays.

Recently, we've started to use arsenic doped wafers and have seen a low rate
(~5%) of the die have their backside metal peel off during the die picking
stage.
The occurence is ONLY on the Arsenic doped wafers (repeat DOE runs have
confirmed this).
Does anyone have #1) a theory on why this is occuring, #2) an engineering
solution for getting around this problem
There seems to be a problem getting an adequate supply of 5" Non arsenic
wafers at a reasonable price, so I'd rather not restrict ourselves from
using arsenic wafers if there may be a solution around the corner.

Currently we are setting up our sputter system to sputter etch the backside
of the wafers just prior to sputtering the TiW thinfilm.

thanks in advance,

Gerry Overton
Process Engineering
i-Stat, Canada