I am using RF magnetron sputtering for deposition of Nb with WO3 film. W target being deposited by DC sputtering and is working fine, while doping Nb and turning on RF power even at very low power i.e 10W, I am facing noise from the supply. I even tried stand alone RF system for W, WO3 and Nb targets, every time I have faced same issue. Returned the RF supply to manufacturer but according to him its working fine, no noise issues . . . what can be the causes ???
How long are you waiting to hold the sputtering chamber at the desired pressure, before turning on the plasma?
Can you clarify what you mean by "noise"?
You always have to use some form of matching network with an RF supply as the RF power behaves a bit like light, and can get reflected by what ever you have at the end of your RF power cable, unless it has some matching system in place, this could be what you are experiencing.
Hi,
Is it a kind of "humming sound" from the supply or power fluctuations in the digital display? I faced similar issues with Al sputtering along with Oxide targets. You can follow "Phil Denby's" comments.
Can you see how much power is "reflected" and how much is actually getting to your target?
Dear Denby, what is the actual reason of back current in RF power supply? And what should be their numerical value at the time of deposition.
Hi Gaurav, Radio frequency power is a strange beast! As I mentioned it is a little like light, if you send it down a cable it may travel down it happily, but then when it gets to the other end it "sees" a transition in the impedance at the junction between the cable and what ever it is attached to.. A bit like light hitting a change in refractive index. If the impedance are very different then the majority of the RF power simply gets reflected and sent back to the power supply.
In order to over come this an "RF matching network" is used at the end of the cable. This typically has 2 capacitors (one in series, one in parallel) and an inductor. The capacitors are remotely tuned using electric motors until you see a minimum of reflected power.
We use a Cesar 136 RF unit with a VM1500AW matching unit, with some fiddling we can get about 300W "real power" and 200W reflected (this is the maximum the supply can cope with and dissipate).
Deposition rates of oxides even at this power are very slow
To complete Phil Denby’s answer, if your matching box (Like a WM1000A) does not automatically find an impedance match, you can open it (of course with generator off) and manually change the number of turns of the inductor coil : this increases the possibilities of finding a good impedance matching.
Normally, it's always possible by adjusting the capacitances and the inductance to find an impedance matching giving a reflected power close to zero.
- Badges
- Science topic
- Similar topics
- Geoscience
- Atmospheric Sciences
- Atmosphere