I have been doing e-beam lithography for only about 2 months, so am a real newbie to the field.

I am trying to write a periodic array of gold cylinders (50-75nm in diameter) and seperated by 40-60nm on silicon (and at sometime in the future on quartz). At the moment I am using a bilayer PMMA (Total thickness 200nm). Everything seems to work, I mean by optical inspection after development, I can see some features, and after metal deposition the feature upon SEM analysis the strucutres are clearly seen, but upon liftoff everything seems to get washed away.....

I guess what I am trying to ask is, has anyone else come across such a problems, and what are the main factors that needs to be kept in mind to avoid this problem.

It is common for me to use 10 nm Ti adhesion layer between noble metal and silicon-containing material.

The other thing to check is your bilayer resist profile, that there is an undercut, so to speak, so that deposited metal on the wafer surface doesn't connect with the metal on the resist.