Hello ,
I have a problem with showerhead pattern on wafer after depositing oxide in STS MULTIPLEX PECVD Chamber.
Process Parameters;
Teos:100 sccm, O2:1000 sccm
Deposition Pressure : 300 mT
RF Power: 300 W
Plate Temperature: 300 C
Showerhead Temperature : 230 C
I have changed the parameters above couple of times to see if this showerhead pattern dissappears, however, I couldn't find any correlation between these parameters and the showerhead pattern.
P.S. I've cleaned the chamber soon, it didn't help to solve this problem.
Thanks in advance
I never used STS MULTIPLEX coater machines but looking at your pictures I have a couple of things… what distance between the shower and the substrate and what deposition rate ? I observed similar effects with different equipments and I found a solution by decreasing the pressure moving to more glow plasmas ( up to 1 Pa ) and the deposition rate. best regards.
Simone Vezzù I concur. Also, might be a good idea to condition the chamber, and mostly the shower head itself.
Lastly, it might be a good practice to flush the chamber with N2 / Ar before igniting the plasma.
I know nothing about that particular coater, but it seems to me that the flow rates are extremely high (only reason I can find to get such a clear showerhead pattern on the substrates). I suggest to lower significantly the flow rates (maybe values around 20sccm or lower) and see what happens. The deposition rate will likely lower, but if you don't get the showerhead pattern at low flow rates, you can than increase them progressively after.
Also, did you check if the showerhead gas outlets are obstructed by coating? It could increase the 'entrance speed' of the gas in the chamber, reaching the substrate more directly instead of being diffused.
That`s an interesting problem.
I notice your deposition plate temperature seems quite low - have you tried increasing this to 350ºC or even 400ºC to see if that helps to improve it - may help the TEOS to flow (along with the other suggestions on conditioning, flow rates). I'm sure the manual should also have some info on chamber conditioning and there should be a BKM process recipe.
Simone Vezzù First of all sorry for my late reply. I have not measured distance between the shower and the substrate as a matter of fact. We have seen that the shower head effect decreases when we increase the pressure and decrease the rf power, but we are trying to understand why it occurs in previous conditions. Inaddition deposition rate is between 50 (Angstrom /second) and 100 (A/s). This rate changes according to the process.
Daniel Poitras First of all sorry for my late reply. We opened the chamber and cleaned the holes, but we have not opened the back of the shower head, so we think that if there is a blockage there, it may be the cause. So if the problem is the blockage in the back, we will try to clean it without removing it. Also we will decrease deposition rate in trial studies and see the situation
Derek Corcoran First of all sorry for my late reply. This is a possibility, but we have always worked at the same temperatures before and there was no such problem, but we will increase the temperature as you said and apply the process that way. I will share the result with you here.
Avi Raveh We will experiment by reducing the oxygen rate and share the result with you.
- Badges
- Science topic
- Similar topics
- Mathematics
- Algebra
- Matrix