[labnetwork] Seeking advice regarding developer/unexposed resist interaction

[Gustavo de Oliveira Luiz]

Thank you everyone for the replies! I am definitely taking note of all of
them.

Some of you asked why I am developing the same sample multiple times. This
is necessary when we are servicing the system and need to test performance
after changes, before moving forward. The idea was to avoid breaking the
wafer and being able to clean and reuse it afterwards. Another use of this
is when adjusting testing overlay alignment, since we need to expose and
develop 2 layers.

I'll try to sumarize below the information that you shared, so it's
available to more people in the network.

   - Some of you confirmed that this is indeed a problem. The plausible
   causes are either the basic nature of the developer, or water — I'd guess
   the former. One of these seems to form a skin on the surface of the resist,
   making it harder to develop.
      - One of you even said that they used this to their advantage,
      producing undercut for lift-off
   - One of you said that even doing the same thing as we do, never saw
   this problem. They said that they use AZ-300MIF (we usually use AZ 400k 1:4
   in our standard processes), and perhaps that makes a difference.
   - The suggested solutions are:
      - Cut the wafer after coating, of course
      - Use pipettes to dispense developer only on the areas of interest.
      In fact, as some of you pointed out, this is the technique used by
      Heidelberg when servicing, we just had no idea why. And,  apparently, we
      should be alright rinsing the whole wafer with water (which makes it more
      plausible that the base in the developer is the cause of resist
hardening).
      - An interesting suggestion was to perform a PEB after the second
      exposure (after one run in the developer) to try and break that surface
      film. We will probably try this one, and perhaps try baking between
      development and exposure as well.

Again, thank you all for sharing your thoughts and experience. It was very
fruitful.

Best regards,
--
Gustavo de Oliveira Luiz, PhD
Applications/Research Specialist
nanoFAB, University of Alberta
+1 (780) 619-1463

On Tue, Jun 15, 2021 at 9:02 AM Gustavo de Oliveira Luiz <
deolivei at ualberta.ca> wrote:

> Hello everyone,
>
> I am having some trouble developing recipes or testing performance on our
> Heidelberg MLA150. Especially when testing performance, it is required for
> us to expose the test pattern once, develop, inspect and repeat. We usually
> coat a larger sample (e.g. a 4" wafer) with resist and expose a small test
> design, allowing us to use the same sample many times. The problem that I
> encounter is that the development time increases after each iteration, and
> it is not due to the developer getting weaker (I tried using a new
> developer to make sure). This behavior is observed on AZ 1500 and AZ 4000
> series resists, with the latter presenting a more drastic change.
>
> It seems to me that the developer changes something on the unexposed
> resist, making it harder to develop. This is made obvious by the fact that
> we don't see this problem when processing multiple samples and developing
> them with the same developer bath. In that case the only problem we
> encounter is the developer getting weaker depending on the exposed area and
> number of samples, as expected.
>
> I thought that it could be over-hydration, since the developer is a water
> solution and we rinse with water as well. But high water content tends to
> make development faster, not slower. Does anyone have more information on
> the interaction between developer and unexposed resist? I'd appreciate any
> input you may have.
>
> Best regards,
> --
> Gustavo de Oliveira Luiz, PhD
> Applications/Research Specialist
> nanoFAB, University of Alberta
> +1 (780) 619-1463
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20210615/2f601ef8/attachment.html>