[labnetwork] Deep anisotropic etching of SiO2

[Robert Nidetz]

Could you mechanically remove most of the SiO2 using something like a
dicing saw and then finish the removal using a wet etch? Maybe create a fin
structure in the SiO2 you do not want and then let a wet etch remove the
fins? Your undercut could be reduced in this manner.

On Wed, Jan 18, 2023 at 8:08 PM Howard Northfield <
Howard.Northfield at uottawa.ca> wrote:

>
> That's a very deep etch.
>
> Regarding dry etching:
> I have recently done etch tests on PEVCD SiO2 with a SAMCO 10NR RIE:
>
> - 50W, 80sccm CF4, 16sccm O2, 3.0Pa: ~21nm/min
>
> - 25W, 25sccm SF6, 10sccm O2, 3.0Pa: ~7nm/min
> - 50W, 25sccm SF6, 10sccm O2, 3.0Pa: ~19nm/min
>
> ... yes for 500um that would be around a 417 hour etch .... not
> feasible/realistic.
> So you can probably rule out these etch recipes.
>
> Regarding wet etching, generally I find it very hard to control and mask
> under-cut is significant.
>
> Howard Northfield
> Research Associate
> Advanced Research Complex (ARC)
> University of Ottawa
>
> ------------------------------
> *From:* labnetwork <labnetwork-bounces at mtl.mit.edu> on behalf of Ningzhi
> Xie <nzxie at uw.edu>
> *Sent:* Wednesday, January 18, 2023 4:23 PM
> *To:* labnetwork at mtl.mit.edu <labnetwork at mtl.mit.edu>
> *Subject:* [labnetwork] Deep anisotropic etching of SiO2
>
> *Attention : courriel externe | external email*
>  Dear college,
>
> We want to perform a highly anisotropic, very deep (~500um) etching of
> SiO2 with a vertical side wall. On top of the SiO2 is nanostructures
> protected by photoresist. The etch depth needs to be controlled with a
> precision of 20% (we are thinking of using another  material as the
> substrate underneath the SiO2 layer, which acts as an etch-stopper). The
> structure and dimensions are shown in the attached image. It would be very
> helpful if anyone has any idea of this kind of etching (either dry and wet
> chemical etch is fine).
>
> Thank you very much.
>
> Best regards,
> Ningzhi Xie
> Department of Electrical and Computer Engineering
> University of Washington
> _______________________________________________
> labnetwork mailing list
> labnetwork at mtl.mit.edu
> https://mtl.mit.edu/mailman/listinfo.cgi/labnetwork
>


-- 
Robert Nidetz, PhD (he/him/his)
University of Michigan
LNF User Liaison and Domain Expert
1241 EECS Building
1301 Beal Ave
Ann Arbor, MI 48109
nidetz at umich.edu
734.496.8065
Book an appointment: https://calendly.com/nidetz
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20230118/b1cd51ff/attachment.html>