[Mitch]

I want to thin out some silicon wafers but leave the atomically smooth side unscathed. The other caveat is I don't want to use HF. If someone has a prep that could do this it would be very appreciated. even if it is an all-day/all-week thing it's okay as long as it can thin out homogeneously. I was thinking spin etching H2SO4/H2O2 unless someone else had a better idea.

[Borek]

Probably unrealistic, but what about dry HCl gas to convert silicon to trichlorosilane?

[excimer]

Unfortunately, HF is probably your best bet here. Piranha doesn't really dissolve silicon well- it just oxidizes the surface, and even it it does dissolve some silicon, it goes bad before making so much as a dent in it. I don't know how thin you were looking to make your silicon wafers, but if you only want to shave off a few hundred nanometers, you could try an e-beam evaporator. HCl gas would probably chlorinate both sides. With HF you could, in theory, just remove the rough side and not the smooth side, as long as you're careful to just expose the rough side to the HF. What I would do, however, is call a silicon supply company and special order thinner wafers and not have to worry about it

Why do you need to thin them anyway?

[Mitch]

What concentration of HF have you used to thin silicon?

[excimer]

I've never done it myself, but I think 30% would do the trick. Higher concentrations will take less time, but it's also.... higher concentrations of HF.

[Borek]

HCl gas would probably chlorinate both sides.

Won't something like parafin block it? Or is applying such thing to polished waffer too risky for the surface?

Besides, the same may happen with HF:

With HF you could, in theory, just remove the rough side and not the smooth side, as long as you're careful to just expose the rough side to the HF.

I am far from suggesting that dry HCl will work, I am just courious why not

[enahs]

I know Potassium Hydroxide can work. A quick search for KOH silicon etching turned up this site:

http://www.mems-exchange.org/catalog/P1454/

[Mitch]

We've tried warm KOH, but it is still way too slow. But, spinning HF just sounds like such a bad idea...

[enahs]

The KOH might have been to slow because of a oxide layer? If so, you can remove the oxide layer with a quick dip in a ~1:20 (by volume) mixture of HF:H₂O (the good stuff). Then the KOH would go faster. According to that site I linked, 20 micrometers an hour. That is 1 mm in 50 hours. How much are you trying to etch? You did say, in your original post after all, that it does not matter even if it is a all week thing

[hmx9123]

KOH does dissolve silicon dioxide.  Very slowly, and not all that much, but it does.  HF is much better.  More dilute solutions work faster, actually (to a point).  HF is a 'big bad' chemical that people don't like using.  The electronics industry uses it all the time, and many non-chemists who have little idea of the dangers of the material use it safely every day.  It is overhyped.  We used to have a squeeze bottle of 49% in the glass shop where we worked.  While I wasn't as cavalier as the old glassblower there who would use it with bare hands, I didn't feel unsafe after using the proper safety precautions.  Wear gloves meant to protect you from it, and have a saturated solution of MgSO4 in a bucket so that you can immerse affected body parts in it if you spill it on yourself.  You usually don't feel the burn until 30 seconds or so after contact, and if dunk your hands or whatever in the mag sulfate solution before that, you won't feel anything.  Respect it, don't fear it, and use it carefully.  Be knowledgeable, not stupid.  It is a chemical that can damage you moderately to severely if you don't respect it, unlike many others that we work with.

[Ψ×Ψ]

Mag sulfate, eh?  I've always heard calcium gluconate.  That's interesting...

[hmx9123]

Calcium gluconate also works.  Any soluble magnesium or calcium salt.

[skippy]

is there anything epsom salt wont do?

[Fleaker]

Wish I had saw this way back when.

I use fluoroboric acid. It works well for silicon.

You can also use ammonium bifluoride (or any alkali fluoride salt) with some concentrated phosphoric acid to make an in situ HF paste.
Paraffin wax works very well for separating what you're etching from what you're preserving.
Avoid using concentrated HF, anything above the usual 48% and it fumes and is quite mobile (and thus dangerous to pour and apply).


Calcium gluoconate in KY Jelly is the usual thing to keep at hand, but I've used benzalkonium chloride solution as well.

[jenik]

I would like to etch silicone wafer over photolitographical mask. I need depth aprox. 20 microns and homogenity less then 1 %.
What is the best chemical for it? HF with HNO3 or something else? What about fluoroboric acid?