We are trying to fabricate free standing thin films (~2um) using SU86002. Since SU86000 is not extensively used in literature, we initially started off with the data sheet parameters and found them to be insufficient since they led to rather brittle SU8 sheets (we spin coat them on copper and then etch away the copper, silicon cannot be used since our final substrate would be a graphene CVD foil on copper). We then looked around and found literature on neural probes that used SU82000.5.
While the parameters used for 2000.5 worked better with 6002, crack formation wasn't completely gone. I suspect it might be due to the differences in thicknesses (500nm vs 2um) however, before proceeding, I wanted to check if there are any significant differences in the two formulations that might spring up unexpected surprises due to the longer hardbake/higher dosage (currently 135mJ/cm2 and 20 minutes at 180 degree celsius with soft/post exposure bakes at 65 and 95 for 2 and 4 minutes).