> I have two visiting students from Germany who are
> making ultrathin SiN force microscope cantilevers,
> of order 350-800 Angstroms thick and up to 200
> microns long.
>
> The last stage of the process is an anisotropic
> etch in KOH to free the cantilevers from the
> underlying silicon, followed by critical point
> drying in C02.
>
> During the etch, the Si bubbles like crazy,
> to the extend that we are surprised that the
> cantilevers even survive!
>
> Survive they do, however we would like
> to suppress the bubbling if possible.  Can
> anyone offer any suggestions?  We apologize
> if this is a well-know problem with a well
> known solution.... it is only well known
> to those who know it well (which is not us!)
>
> Best wishes from a fan of MEMS ... John Sidles
The reaction (from a paper by H. Seidel in J. Electrochemical Soc.) is
Si + 2 OH- + 2 H2O -> SiO2(OH)2-- + 2 H2
The gas bubbles are hydrogen.

I, too, have had trouble with silicon nitride membranes being broken by
bubbling underneath.

The etch rate of Si in KOH slows with temperature. It also slows with
increasing AND with decreasing concentration around a concentration of about
18% by weight.

In the future, to avoid this the outgassing problem, I will try using a lower
etch temperature to slow the etch and thus the H2 formation. Going to a lower
KOH concentration than 18% will also slow the etch, but this also leads to
rougher etch surfaces.

--Kirt Williams
Berkeley Sensor & Actuator Center