durusmail: mems-talk: RE: mask etching for glass
RE: mask etching for glass
RE: mask etching for glass
Marc-Alexis Gretillat
1996-07-24
> Dear colleagues,
>
> I am trying to isotropically etch glass wafers. I want to etch the mask
> directly, but after etch in 1:1 BOE/HF, I found that the Cr was lift off
> and etched away, and the lateral etch rate is much higher than vertical.
> I don't understand why is that, maybe because the etching of Cr, but accord-
> ing to CRC hand book, Cr should be able to stand for HF. Is there anyone did
> the similar process? How can I get good isotropic etching of glass wafer?
> Any suggestions and explanations are highly appreciated. Thanks a lot.
>
> Ingrid Y. Xu
> Microfabrication and Application Lab
> The University of Illinois at Chicago
> TEL: (312)-413-7576
>
********************************************************************************
Dear Colleagues,
We have also worked on this problem at the Institute of Microtechnology,
University of Neuchatel Switzerland.
Effectively Cr layers have a relatively poor adhesion on glass. Former
work by us where using a Cr/Au layer to etch glass. But this technique
requires to etch gold and to etch the chrome layer in two different
solutions and a lot of pin holes were present limiting thus the etch to
20 microns.
We decided to change our masking material and made sevral tests until we
found that an amorphous silicon can be used for such a purpose. We also
found some works made at the TU Deft in the Netherland reporting the use
of LPCVD Polysilicon to etch pyrex glass as well as work at the MIT
Cambridge USA where poly masks were used to etch quartz.
As the polysilicon can be batch deposited, we usually used it now to
etch pyrex #7740 glass wafers.
Nevertheless, both (amorphous silicon and polysilicon) are etched in a
plasma etcher to define the mask. And with such layers it has been
possible to etch through pyrex #7740 glass wafers of 0.5 mm. It means an
etch depth of 500 microns without any problems. It is clear that this
etching technique is completely isotropic due to the use of the HF
solution.
This work will be presented in Eurosensor '96 at Leuven Belgium and in
micro-TAS in Basel switzerland, applied to the realisation of micr
capillary tubes.

Marc-Alexis Gretillat
Institute of Microtechnology
University of Neuchatel
Rue Jaquet-Droz 1
P.O. Box 3
CH-2007 Neuchatel
Switzerland

E-mail: marc.gretillat@imt.unine.ch

present E-mail adress until August 24 1996: gretilla@mtl.mit.edu

********************************************************************************
**


reply