PDMS membranes are very tricky to handle. Trying to peel could result in tears or membrane simply holding onto itself (which makes it very difficult to untangle). I guess the best method of handling it would depend on your what you are going to do next with the membrane. At this point I would suggest 2 things:

1. Once membrane is spun and cured oxidize its surface in the plasma oven (O2 or air plasma) and bond to a larger PDMS piece for handling (drum-like structure).

2. Simply peel the membrane under water (do it gently). This way you will be able to fix any tangles and the membrane will float on the surface. You can also use a sacrificial layer (lift-off approach) of high molecular weight PVA (poly-vinyl alcohol), which would dissolve in water as you are peeling PDMS of.

Yes i would definitely follow dmitry to peel the pdms under water.

Yet there is another way to fabricated very thin pdms membranes using a reliable, low tech & low cost approach.

Take a glass plate (microscope slide) of the desire area. depending upon the thickness of the required membrane attach a thin duct tape (self adhesive tape) on both the extreme end surface of the slide. Apply a very thin layer of some hair oil manually on the slide. Now pour slowly the pre mixed pdms on to this slide. Take another glass slide and apply the hair oil. Similar to the process of preparing the microscope slide without capturing the air bubble in it, start laying the second slide on the top of the pdms from one end very slowly. Let the pdms adhere to this on coming slide by itseld as you lower the slide. Then let the slide rest on the pdms surface for a while. the excess pdms will be pushed out. Put some weight on this slide without moving it. keep it to cure at the set temperature. The sandwiching technique to obtain the thin pdms membranes is very efficient.

The oil in this procedure helps to reduce strong adhesion of pdms on to the glass surface. The oil can be easily cleaned by dipping the membrane in ethanol and slowly scrubbing the membrane with tissue. In case the oil is not used it still gives good results. I was able to produce pdms membranes of lengths 5cm, widths 3cm and thickness down upto 50 microns. Very the number of duct tape layers to get the desire membrane thickness. Or even coverslip alongwith duct tape can be use in similar manner to create the desire thickness. The procedure works efficiently on any surface. Once the process is setup, it can produce sufficient amount of membranes in very short time.

Take thin (e.g. 25 micron) Cu foil. Spin coat PDMS on it up to desired thickness. Cure PDMS. Place PDMS/Cu sandwich into Cu etching solution (Cu immersed in solution). Capillary forces will keep the sandwich at the surface. Cu will be etched and PDMS will float at the surface. Scoop PDMS membrane using appropriate wire loop or substrate and place it to water for cleaning.

Try using a thin film of aluminum foil as a sacrificial layer between the kapton and PDMS. Aluminum foil has a low surface energy that will assist easy removal of PDMS.

You should release the PDMS membrane from Kapton by immersing the laminate in hot (say, 50C) water. It will detach by itself - you do not need to peel.