Hi Andrea,

Before trying to make this be sure to spend an hour with a structural
mechanics simulation to determine how much the structure will bend
simply due to gravity. I have a very long (2000µm), thin (1µm)
cantilever, which will drop to the floor below if not hung vertically.
With a 2000/1 ratio, the drop is on the order a couple of thicknesses.
Comparing to your numbers it seems you're in the same ballpark. I'm
guessing you will need support pillars to avoid this.

Of course, in addition to this, it might be very hard indeed to perform
such a large underetched area using wet etching. Perhaps gas-phase
etching might be an alternative?

Best of luck
// Morten

Andrea Mazzolari wrote:
> Hi all,
> i need to realize a Silicon On Nothing (SON) structure of lateral size
> about 0.5x0.5mm. Silicon layer should be 20nm, and nothing layer in the
> range 40-60nm.
>
> Some questions:
> 1) which is the best agent to etch SiGe and preserve Si ? (I've found that
> PA:HF:H2O, 1:1:1 where PA=paracetic acid) should have a good selectivity,
> but did not found a numerical estimate.
> 2) is it possible to realize a so large SON structure or the thin silicon
> layer will surely destroy ?
> 3) which will be dislocation density in the thin silicon layer ?
>
> Best regards,
> Andrea