Solvent casting polyethersulfone (PES)

Nobody knows how to do this?

Thanks,

Phil Hobbs

Phil Hobbs wrote:

You want the solvent to evaporate slowly, thereby allowing internal stresses in the polymer to relax as much as possible and thereby preventing cracking. Criteria for slow solvent evaporation can be somewhat tricky, as it depends not only on the inherent evaporation rate of the polymer itself, but also on the specific interactions between the polymer and the solvent - some solvents will really cling tightly to the polymer. Consider using two or more solvents that are compatible with each other, or get wild and crazy with using a nonsolvent that is mixes well with a solvent.

More solvent can help with better stress relaxation but also leads to slower production rates.

Heat can also be used to relax internal stresses, but increases evaporation rates.

There is a good deal of art to this.

John Aspen Research, -

Opinions expressed herein are my own and may not represent those of my employer.

What is your substrate? Cooling rate between the cast film and the substrate can be problematic. An example is the SU-8 epoxy resist which requires attention to the polymer substrate interface. Rather than placing the substrate on a high surface area cooling block we use a fluroware spider to minimize the surface contact area.

boer

Thanks for the responses. We've tried n-methyl pyrrolidone (NMP), gamma butyrolactone (GBL), tetrahydrofuran, and a couple of other things. GBL seems to work best, but a 10% solution of PES in GBL turns to hair gel in a couple of days. GBL evaporates at a lower temperature than NMP.

We've tried baking at two different temperatures, e.g. 80 and 200 C. The films want to crack even coming off the 80C hotplate.

N > What is your substrate? Cooling rate between the cast film and the substrate can be problematic. > An example is the SU-8 epoxy resist which requires attention to the polymer substrate interface. Rather than placing the substrate on a high surface area cooling block we use a fluroware spider to minimize the surface contact area.

The substrate is silicon, which can't be changed, since it's the silicon we care about in the finished device. The CTE mismatch is quite large, but other polymers (e.g. photoresist) don't show the same cracking problem. The temperature of the solution when we spin it on seems to matter considerably. Our hope is that the film can handle the CTE mismatch by itself, and that it's just the combination of the shrinkage on drying plus the CTE mismatch that's the problem.

There exist solvents for this stuff that we can evaporate at 250C, but there are a couple of problems: first, they're nasty toxic things, and second, we need the outgassing rate of the finished film to be very low at 200C--it'll blister the overlying layers during litho otherwise.

Cheers,

Phil Hobbs

You need to prevent water vapour from condensing on the polymer solution while the solvent is evaporating, mainly when using solvents that can also dissolve water (gamma butyrolactone, tetrahydrofuran, NMP are all of this type). The water is left trapped in the matrix, making it brittle. If you use a solvent that doesn't attract water much, some water will still get in from untreated air, but it produces just a whitish bloom in the surface. Ideally, you would do the operation in a dry atmosphere.

In the absence of that, I like chlorinated solvents! (But I concede they are a bit toxic and you need good vapour extraction.) 1,2-dichloroethane should dissolve your polymer and not attract any water. (I use this for solvent welding PMMA.) A little higher boiling co-solvent, such as NMP or GBL, should help to reduce stresses in the final stages of drying, when the film is less porous to water vapour.

Thanks, that tallies with our observations about film quality vs. humidity.

We'll try being more careful about water vapour.

Cheers,

Phil Hobbs