Thermal reflow of +ve resist 3612

jim kruger jimkruger at
Tue Dec 12 09:45:49 PST 2006

I do not have a complete recipe for you, but I have
found that about 15 minutes in the EEprom eraser (the
old DUV cure sytem near EV-Align) followed by about 5
minutes on a 200C hotplate will cross-link 3612
without too much change in small features.  Larger
features (the edge or corner of a big block of resist)
show some fringes, indicating a change of thickness at
the edge, but I believe the resist edge at the
substrate is not moving.  Adjusting the UV time and
the hotplate temperature might give the smoothing you
  I use this recipe to prevent mixing of a second
layer of resist so I can add patterns with a different
mask.  The cured resist will withstand a short (few
second spray) of acetone, I have not tried a long
acetone soak. It is OK in the developer for the second

Another technique for edge smoothing is to use a low
pressure O2 plasma descum to remove 5 or 10% of the
resist selectively to the substrate before the
substrate etch.  This can be combined with the UV +
hotplate as modified from the recipe above.

What are you etching in which system/recipe? 
Selectivity to resist?

--- "Rohan D. Kekatpure" <rohank at> wrote:

> Dear Labmembers,
> I am using a process in which I expose circular
> patterns in the 1 um thick
> 3612 resist using Nikon stepper. I am concerned
> about the circularity and
> the line edge roughness in the resist (which I
> intend to use as a dry
> etch mask). Therefore, I am wondering if anyone has
> tried a reflow on this
> particular resist. If indeed you have tried, it
> would be great if you
> could provide me the starting parameters (reflow
> temperature and time).
> Thanks in advance for your help.
> -Rohan

Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 

More information about the labmembers mailing list