Continuing the discussion from Metal Liftoff not working – metal film stays everywhere on the resist:
This Topic seems to come up a lot, so I am making a separate tech note here with some images.
Problem statement: After doing metal deposition, there is sometimes poor liftoff results. In this scenario, most often all of the metal deposited does not lift off, or if it does, the edges of the metal or not clean (sometime ragged)
Solution: Using a Bi-Layer resist structure. this is where you use two different resists that have different solvability in the developer, so that there is an “undercut” created, facilitating easier lift off.
In this example we will talk about PMMA (and E-beam resist) but this methodology also works with photoresists, using LOR (Lift-Off Resist as an underlayer)
- Begin with your bare substrate
- Spin coat with the lower molecular weight PMMA (eg: PMMA 495) – This thickness should be at least 1.5x-2x the thickness of the metal that will be deposited. Bake this.
- Then spin coat the higher molecular weight PMMA (eg: PMMA 950) and bake that
- Expose on EBL
- Develop the resist – the lower weight PMMA will develop away faster resulting in an undercut below the higher molecular weight PMMA
- When doing E-beam deposition, the metal will now be isolated from the resist, and will not fuse before liftoff.
- Nice structures left behind!
Acknowledgement: This picture is modified from one of 30 PPT slides from the MNFC Cleanroom lab at Princeton University – Very good resource. Find the full presentation at the following link