700nm~1um thickness photoresist gasket

There are four main challenges for making such a thin photoresist gasket on a little chip.

Q1: How can I get under 1um thickness photoresist gasket?

Q2: Is the adhesion of photoresist gasket strong enough for further usage?

Q3: How can I write the pattern onto the gasket that will have the center window open?

Q4: How can I make sure that no photoresist will clog in the pores?

 

Q1: we must make a photoresist layer that is thin enough to reach our requirement. I use SU8-2002 and spin coat it on a sample chip.

Image1-1(up): I use double-sided tapes to hold the chip because the chip is too small to be held by the vacuum.

Image1-2(down): The chip after spin coating. This one is successful on the uniformity.

The spinning setting is 500rpm for 5 seconds; 800rpm for 10 seconds (These two steps are aiming for more uniform surface); 9000rpm for 60 seconds (This step is to thinner the layer).

Image2: I wiped off half of the gasket in order to measure the thickness of the photoresist gasket.

The profile of the surface is at below:

Image3: Part of the surface profile. The peak is the accumulation that happened when I wiped off half of the gasket. The thickness stables at around 900nm. (10000 angstrom = 1 micron)

Image4: This is a very beautiful and flat graph. We can clearly see that the thickness nearly stables around 700nm. (10000 angstrom = 1 micron)

 

Q2: Then I proceed to next question, which is to test the adhesion of this thin photoresist gasket. The reason I choose SU8 as photoresist is that it has a good adhesion in most of time.

The procedure is:

Spin coat the chip; soft bake at 95 degree C for 1 minute;

Put HMDS (adhesion promotor) on the glass plate.

Bring the chip and sample glass together; soft bake at 95 degree C for 90 seconds;

hard bake at 185 degree C for 20 minutes.

The adhesion is not as strong as PDMS, but it is strong enough to stay together if you don’t mean to separate them. Here is a pictures of one sample (this is not the perfect one):

 

Q3: There is a new laser writer in our clean room that can write any pattern on any photoresist layers. So, I am trying to write an open window at the center of chip. Like the picture illustrated below:

I’ll use laser writer to expose the blue part, then bake it at 95 degree C for 20 minutes. After that I will use SU8 developer to wash away the white part (unexposed). However, SU8 has a very low absorption rate for the laser writer we have. After failing several times, I finally successfully pattern the chip like the picture showing below:

Image5: Sample chips after post-exposure bake, before going through developer. You can clearly see the square at the center of the chip.

Image6: Sample chip after develop. The center square is clean now.

Then I continue to test the adhesion of this photoresist gasket that has been exposed. It is the same procedure as Q2 illustrates.

Image7: Exposed chip bonded stably with the glass plate. This means the exposure step won’t affect the adhesion of the photoresist gasket.

 

Q4: Not done yet.

 

Summary:

Now we have approved that making a 700nm thickness photoresist gasket is possible. However, the laser beam from the laser writer I am using is the inappropriate one for exposing SU8. The absorption rate is too low to expose the whole surface completely and nicely. Since we don’t have a chance to change a new laser beam, I am looking for a new photoresist material to replace SU8. I will continue working on this project this fall semester. And I believe finding a substitute for SU8 is not as hard as the progress I’ve made until now.