Hello,

I use a combination of wet etch (HNA, very fast) and dry etch (XeF2,
high selectivity) to etch the back side silicon of SOI wafers.

I mount my wafer, device side down, onto a sapphire carrier wafer using
Apiezon Wax W. At this point, it is important to make sure that the
sides are covered well with wax. The Si etch rate in HNA is very high,
about ~10um/min. After about 25-30 minutes, when I start to see oxide at
the edges, I stop the wet etch and etch the remaining Si using XeF2.
XeF2 has a very high selectivity over oxide, so you get a nice etch stop.

After all this, I use Trichloroethylene or Chlorobenzene to dissolve the
wax and release the devices.

I hope this helps.

Emre

PS: Here is the HNA recipe I use:

HNA =  (25wt.% HF (~49% conc.) + 35wt.HNO3 (~70% conc.) + 40wt.% HAc
(glacial))


Om Suwal wrote:
> Dear Kagan,
>
> In my experience, the use of thermally grown oxide does stand well as etch
> mask for both TMAH . The oxide etch rate is ~2A/min for TMAH @80C. Thermally
> grown >3000 A oxide is sufficient as etch mask to etch for 500 um
> Si substrate of SOI.
>
> Regards
> Om
>
> On Thu, May 6, 2010 at 3:44 AM, Kagan Topalli
 wrote:
>
>
>> Hi,
>>
>> PECVD nitride and oxides cannot withstand several hours of KOH or TMAH
>> etching. LPCVD nitride or thermally grown oxide might be the choices for
>> masking KOH or TMAH etching.
>>
>> Regards.
>> Kagan Topalli


Mustafa Emre Karagozler
Graduate Student
Carnegie Mellon University
Department of Electrical & Computer Engineering
5000 Forbes Avenue
Pittsburgh, PA 15213
http://www.andrew.cmu.edu/user/mkaragoz/