I am trying to fabricate microthruster on 6 inch Silicon wafer, which consist of chamber, channel and nozzle. When i am etching Si by KOH etchant, the channel area is merging with nozzle and chamber due to convex corners. I would like to etch nozzle and chamber first, then channel. Can any one suggest me etchant for channel area. After etching it should don't merge with rest areas.
Thanks in advance.
For vertical channel etching the RIE (reactive ion etching) is the preferred procedure.
We have tried DRIE etching but after that Lithography pattering is problematic. I would like to etch channel only.
Can you provide more information regarding your process? Is there any way you can do the DRIE as the last step for your processing on the wafer? That way you can do all your other processing without the drastic topography from the DRIE which makes lithography problematic.
On the other hand, if you can get access to a photoresist spray coater which will not only coat your top surface but the sidewalls of your DRIE channels, then you should be able to do photolithography after the DRIE.
Sir,
Unfortunately we don't have spray coater. On 6 inch wafer Wafer target mark, Pad oxide, LPCVD, PECVD oxide, KOH pattern, Etching, Nitride removal. These steps are involved in the process. When I am etching with KOH all parts are merging due to convex corners. If etch channel area at the end of process, Photo resist can't with stand up to 100 micron silicon etch. We are using spin coaters. So we can't pattern on uneven surfaces.
After KOH, you do etching. What is it you are etching? Can you etch and then do DRIE?
Photoresist should be able to hold up to a 100 micron DRIE, but I think it depends on your resist and what tool you are using.
For silicon etch, SPTS si etcher. KOH pattern then Silicon etching.After KoH pattern and etching of Silicon we can't coat with PR as we don't have spray coater. We are using positive PR.
Have you thought about using a thick high-aspect ratio photoresist for subsequent lithography steps? I don't know your geometry constraints, but a resist like AZ 40XT can coat to 100 micron thickness and resolve features to about 25 micron at that thickness (better with better process control). If you used a resist such as this, you could move an etch using DRIE earlier in your process. There's some tricks associated with thick resist processing that I can detail to you if you want to go down that path. Send me an email and I'll send you a sample process traveler.
Alternatively, you could be a little more rigorous with the geometry defined on the mask for your silicon etch. If you don't mind a little more fabrication complexity, you can KOH or EDP etch a series of rotating lines to determine the exact orientation of your silicon wafer (wafers are usually a little miscut) and then align all steps to that. Then, you can design your mask with geometrically compensated structures that etch properly given the crystallographic orientation of the silicon.
Personally, I would try using a thick positive tone resist if your geometry is large enough to accommodate the coarser resolution imposed by the thick film.
Can you try DRIE from the backside? You should be able to mount your KOH etched wafer onto a carrier wafer using something like parylene and then be able to do the lithography on the backside since it will still be planar. The carrier wafer will also allow enable the wafer to safely mount in the DRIE machine. Parylene cleans up pretty easily, but you'll need to be careful since it will liquify during your soft bake steps and can make a mess.
As Jere recommended, use a thicker PR for the DRIE. This will enable you to finish your 100 micron etch. In the case of mounting your wafer, you'll likely need extra photoresist since the heat transfer will be reduced and your selectivity will decrease.
Based on your statements, the problems sounds like design not etchant. You need to include convex corner compensation in your etch mask design. If you search for info, you should find hundreds of articles about it.
Do not use EDP, it is too toxic. If the mask (oxide?) does not survive the etching, you can switch to a nitride mask or use TMAH etchant.
The concentration of KOH and etch temp will affect the etch rate of high order planes so you need to make sure you mix your etch solution carefully. never use pellets. Pellets absorb water from the air and it is hard to get the right concentration. Make sure you don't lose water to evaporation or keep topping up with water.
Best of luck with project.
Thank you for the valuable comments to all.
Afetr etching the nozzle and chamber area by KOH i would like to etch channel by XeF2 etchant. as the mask can be nitride for both. Does any body has experience of XeF2 etchant.
What is the etching rate of AZ40XT? Is it etched only by oxygen or also by Sf6?
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis