> I have a Ge substrate on which 200 nm SiO2 and top of that
> 200 nm SIN film was
> deposited ( PECVD). After annealing at high temperatures,  HF
> etch does not
> get rid of the deposited nitride layer though it does uniformly before
> annealing. I have tried etching for more than 15 min in HF (48 %).

The etch rate of PECVD nitride in HF drops after annealing,
as least partially due to outgassing of hydrogen.
As an example, the etch rate of stoichiometric silicon nitride
(which has much less hydrogen) in 49% HF is 140 A/min.
Thus, in your case, a longer HF is probably needed.

Alternatively, you can use 160 C phosphoric acid,
but this may be even slower (28 A/min for stoichiometric nitride).

        --Kirt Williams Agilent Technologies