KOH - backside etch - frontside protection

[ChrisPark]

Hi all,

I'm planning on a process where I want to create membranes in a KOH etch from the back side but the front-side is already structured and must remain dry. I want to control the membrane thickness (~18µ) optically. Does anyone
have experience with such a process or knowledge on the equipment I would need?

Thanks!!

 

[Colbhaidh]

You will need front side (and back side is masking required) polyimide spin on coater and cure oven.
You will need a solvent bench to remove polyimide and possibly a oxygen plasma resist strip system.
Do you need to remove backside silicon and stop 18um from the surface? Without an etch stop this will be difficult to do accurately with just KOH.
There are semiconductor step prifiler measurement systems that use a probe needle to detect step heights which would work. They are expensive but available on the 2nd hand market.
Another less accurate way is that KOH will etch along the crystal plane so you can measure the width of the slope under a microscope and calculate the depth.

 

[IRstuff]

A further comment to the crystal plane preference, it also means that you'll lose roughly the thickness of the wafer laterally because of the etch slope.

So, a 0x0 um finished feature on the frontside will occupy about 500x500um on the backside, assuming a 19 mil etch depth. This can result in structural problems with the circuitry if you haven't allowed for that. Flexing of the membrane can could generate unforeseen leakage currents in the surrounding circuitry.

TTFN

FAQ731-376

 

[ChrisPark]

Thanks guys, I found a company which provides mechanical fixtures for wafers. These fixtures keep one side dry while exposing the other to KOH. They also offer systems for electrochemical etch-stop to control the membrane thickness.
(check ammt.com)