|
|
Ask the Staff
john.d.watson.2
2012-03-15 05:30 PM
Determining photoresist exposure
I'm working on setting up a process flow for etching a mesa and evaporating ohmic contacts, and I've got a question about how to determine the exposure time for different photoresists apart from "the last guy in the group told me this worked." The manufacturer's specs I have found online seem to call for much lower exposures than what other people (including people who are good at processing) have told me. What is the trade-off between over-exposing or under-exposing for etching and lift-off processes? As a specific example, for AZ1518 the datasheet lists 339 mJ/cm^2 at g-line as the "Dose to print". Is it safe to assume that 34 seconds at 10 mW/cm^2 with the aligner set to the g-line will correctly expose the resist? Thanks!
michael.d.courtney.1
2012-03-16 10:03 AM
Photo resist exposure
Keep in mind that we set up the mask aligner exposures at 10mW/cm2 on the h-line wavelength at 405nm. If you look at the mercury arc lamp spectrum, you will find a larger peak on the g-line at 436nm then what is found on the 405nm h-line. Since we do not use exposure filters, the substrate receives the entire spectral response of the mercury arc lamp. If the AZ1518 photoresist manufacturer data lists a dose of 339 mj/cm2 for a g-line exposure, your exposure times will likely be different because of the unknown value of the g-line intensity. Our Staff Scientist Aamer Mahmood, indicated that he had used 34 second exposures with AZ1518 but at a higher intensity. He also indicated that the amount of substrate reflectance will also cause a difference with your exposure time. Your developer concentration and develop time is also critical for proper critical dimension control. I suggest that you set up an exposure matrix and measure your critical dimensions on your substrate and see how that compares to your photomask CD's.
|
