Hi Ping, I agree with Bob, you are likely getting some sputtering of your mask layer followed by micromasking. Do you have a capacitive or inductive RIE? Can you monitor the DC bias of your plasma? Here's some things to try: 1) Change masking layer 2) Sounds like you're having problems with reproducability that is ultimately effecting the DC bias of the plasma, i.e. ions are acquiring enough energy to sputter the aluminum. First, be sure your electrode temperature (the chiller) is not below room temperature. A cold electrode will condense water vapor and change the plasma chemistry. This effect MIGHT explain why your second sample looked much better. Second, run an oxygen plasma at high power density and high pressure (say 200-400 mTorr) before you begin your process. This will remove all organic contamination. 3) If you still have problems reduce the ion energy by either increasing the process pressure or decreasing the power. Good luck, Mike >>> ping@virginia.edu 10/8/2003 2:34:13 PM >>> Hi All, We had very serious problems with RIE grass when we were etching chips. In the chip, we used Al as mask for 7um SiO2 anisotropic RIE etch (CHF3 and O2) and we tried to put the chip on steel plate, Si wafer or quartz to avoid the RIE grass. But there is not much difference. And we found that if we etch 2 chips in sequence, then the first chip etched has much grass, but the second one is very clean. Recently, there was leak in the RIE system and we fixed it, but etching chips in sequence to avoid grass doesn't work and the result seems much worse than before. Does anyone have any idea about why we can get a better result for the 2nd chip (all the parameters are same) and why it doesn't work after fixing the machine? Also if we just etch 1 chip, sometimes it has lots of grass, but sometimes it has less grass and is acceptable (open area without Al mask has more grass than small gaps). The result is not stable, and varies without reason. Does anyone have idea about this? Cleaning the system doesn't seem to help much. Last, because the system is not reliable and produces grass without forewarning, is there any kind of method that can tell if the RIE system is good to run? I mean, is there a method that can test if we will get grass in the etched chip before we etch it except using our chips for the test every time, because the number of chips is limited and we don't want to waste them. And do you have any good idea about avoiding the RIE grass? Thanks a lot for your help. Jiaping _______________________________________________ MEMS-talk@memsnet.org mailing list: to unsubscribe or change your list options, visit http://mail.mems-exchange.org/mailman/listinfo/mems-talk Hosted by the MEMS Exchange, providers of MEMS processing services. Visit us at http://www.memsnet.org/