Eric -

a couple months ago  I was trying to solve the same problem  - I was
searching  a substitute  for CVD nitride in KOH etch process.

My best finding was -  TiO2 obtained by thermal oxidation of Ti.
Ti metal was evaporated using a regular vacuum evaporator equipped with
either resistance or e-gun heaters.  Thermal oxidation was performed simply
on a hot plate at 500 C for 10-15 min. in air.

The thickness of Ti was 50-100 nm. The adhesion and the resistance of the
thermally oxidized Ti in KOH was perfect.

If you need to pattern the mask for KOH it is easier to pattern Ti first by
wet etching and then oxidize it (rather than etch TiO2).

I would be very interested to know if these observations are
confirmed/reproduced by someone  else in his/her lab.

Nickolay


P.S.: If you have access to PECVD depostion of nitride, I can can give you
a reciepe how to make PECVD nitride absolutely resistant  to KOH.


hugo wrote:

> Dear Sir
>   I'm one of Material Science student in Taiwan . Now I do have a
> serious problem in etching my specimen ,so it would be great if you can
> do me a fever . Here's my parameter below:
> 1.) 8 inch wafer ( thickness 750 micrometer )
> 2.) LPCVD 2000A  Silicon Rich Oxide ( SiOx  ; x<2 )
> 3.) After several process I have to remove back Si to form a SRO
> membrane
>  first :I spin PPR on my wafer , prebake... exposure .... develop
> .....baking
> second : sputter metal to be the mask
> Third : Lift off PPR=
> Fourth : Etch @ 80 oC KOH ( 20% wt )
>
> As I know Si3N4 was the best choice to be the etching mask , but my lab
> can't CVD Si3N4 . So could you tell me which metal can be the mask . I
> had been sputter  Cu and Ti but they all failure during etching .
> thanks
> Best regards
>  Eric Chang