Photolithography help for small features (2x2 micron boxes separated b

[goforhenry]

Im currently working on a project where I have small features 2 x 2 micron boxes separated by 2-4 micron trenches in every direction. my current process is as follows

Piranha Clean wafers for 15 minutes

Dehydration at 150 C for 10 minutes

I spin coat the wafer with hmds at 2000 rpm for 60 seconds and soft bake at 150 C for 60 seconds

Using S1813 G2 photo resist I spin at 4000 rpm for 30 seconds and soft bake at 115 C for 60 seconds

Then I use g line (436) With an intensity of 10 and a dose of 155 mj/cm^2

Develop using developer Mf-319

The issue im having is that it seems that anything past 45 seconds my photoresist features will start to lift off. According to the website (Dupont) they were able to develop small features (lines smaller than 1 micron) with no issues. Their process is as follows,

Using photoresist S1813 G2

coat wafer with 12300 angstroms of photo resist

soft bake at 115 C for 60 Seconds

G line 150 mj/cm^2

develop doing 15 + 50 sec double spray puddle

Besides a variation in the thickness of the photoresist on the wafer and the use of hmds we follow the same recipe although I get different results. The other issue I see is that when I put my wafer in the developer I submerge the wafer for 45 seconds shaking the wafer every ten seconds. but ill have issues where I get a lot of newton rings in between my 2 x 2 micron boxes. when I developed it for 45 seconds I got the best results although for some of the wafers it seems that the developer will start to eat away at the middle of the boxes giving me a "donut" shape. Any help would be appreciated

 

[Tom.G]

Just throwing an idea against the wall here to see if it will stick.

Is there any way to put an image sensor in place of the wafer to check for a problem with the mask or optics?