The authoritative public forum
  for Metal Finishing since 1989
-----

Etching 2 micron features into stainless steel

January 6, 2011

Hello all,
I am trying to etch channels into stainless steel that are 2 micrometers (~8x10^-5 in) deep and 2 micrometers wide. I have used 40% ferric chloride ⇦this on eBay or Amazon [affil links] in 1M HCl with some success, but I can't get the features deep enough without too much undercutting.

I know these kinds of features are much smaller than what most people here work with, but since there aren't too many scientific papers on the subject, I don't know where else to turn. Any tips or ideas would be of great help.

Ken Pradel
Researcher - Chicago, Illinois, USA

---

Ken

Consider electropolishing. Monitor the micro-amp-seconds to control the material removal.

Willie Alexander
- Colorado Springs, Colorado
First of three simultaneous responses -- January 6, 2011

I will offer an uneducated guess that a search of electro etch might provide a way of getting less undercutting as well as being faster.

James Watts
- Navarre, Florida
Second of three simultaneous responses -- January 6, 2011

With any wet chemical etching process you won't be able to avoid the undercut and, as you mention these dimensions are very small. I would suggest to look at technologies as Ion etching or ion beam etching,which should be able to provide better and controlled results.

Marcus Hahn
- Zug, Switzerland
Third of three simultaneous responses -- January 7, 2011

---

If you're etching by immersion, a change to spray etching directed perpendicular to the surface may reduce undercutting.

Jeffrey Holmes, CEF
Spartanburg, South Carolina
January 7, 2011

---

January 10, 2011

Hi Ken

I must disagree with some of the contributions. First 2 micron features are very large compared with those routinely achieved in microelectronics. And secondly, ferric chloride ⇦this on eBay or Amazon [affil links] is the standard etch for stainless steel at these dimensions. Electroetch works like an electrocleaner and will lift the edges of the resist.

I assume that you are using a photoresist, although you do not say.

The first consideration is good resist adhesion and the surface must be adequately prepared; alkaline soak clean, dilute nitric dip with good rinsing.

Resist thickness must be adequate and controlled. Resist exposure must be correct; either established by a test pattern or better by an exposure test wedge (your resist supplier should advise. Develop and dry - probably low temp. bake particularly if using a solvent type resist.

To obtain a good edge definition to the resist a collimated (parallel) UV light source is needed.

Now the undercut. Undercutting is unavoidable but it can be minimised. Immersing in an unagitated etch is the worst option. Directional (spray) etching is much more directional but requires adequate resist thickness and strength to withstand the mechanical forces imposed.

It is normal practice to measure the undercut (etch factor) and allow for this in the original image. Try starting with 1.5 micron image and etch 2 micron deep but do not expect to obtain any right angles.

If a square section feature is important you might consider an alternative reverse technique. Apply 2+ microns thickness of resist with a 2 micron pattern and plate with Ni, Cu etc.

Geoff Smith
Hampshire, England