He O-ring in PlasmaQuest

Jim McVittie mcvittie at snf.stanford.edu
Tue Nov 23 10:39:22 PST 2004


Thanks for your idea about a blockage in the He line. I agree the
results are consistent with a blockage. If Cesar or Elmer can not show
that your idea is wrong or find a alternative reason for the low flow,
it looks like we are going to have to take the electrode feedthur apart.
A likely blockage would be a kink in the teflon line.

As for the 0-ring policy, it was agreed by the GaAs faculty sometime
back that we would support higher temperatrure (150 to 200 C) InP
etching as long as it did not adversely impact the present etching
processes. To achieve these high temperatures and to continue operating
at the standard lower temperatures, it is essential that we use a Kalrez
o-ring. Yes, we will be going back to Kalrez. Note, the tool originally
came with a Kalrez o-ring. Since we were not operating at the higher
temperatures requiring Kalrez we switchg to the much lower cost silicone
o-ring material.

Regrading what materials are allowed to be etched at the higher
temperatures. I have not given much thought this to this issue. I will
have to get back to you on this. I do not want to open the tool up to
processes which would adversely impact current processes.


jim kruger wrote:

> What is the policy on wafer chuck O-ring material (and
> hence chuck temperature limit) in PlasmaQuest?   I
> assume 120 C is the limit for silicone O-rings and 220
> or 250 C for the Kalrez.  I am interested in high
> temperature operation and I believe Hatice Altug is as
> well.  I use the chiller cooling up to 90 C  (chiller
> at 80 C) but switch to the N2 gas cooling for higher
> temperatures.
> I see that the O-ring material was changed back to
> silicone last night; I encourage use of the higher
> temperature Kalrez O-ring if it is feasible.  What
> will the policy be?
> I see that the new O-ring last night appears to have
> been overheated already.  This would be consistent
> with a blocked or restricted He line to the wafer
> chuck.  On Sunday, 11/21/04, I noticed (and sent a
> comment) that the He flow at 10 Torr had dropped back
> to a more normal 6 sccm but that with no wafer, the
> flow was only 8 sccm.  This seems wrong to me since
> 16 or even 25 sccm He flow at 10 Torr was recently
> reported.  I was using the chiller at 80C with the
> chuck at 90 C.
>    I will check “no wafer” flow as part of “system
> normal” checks on start up and I encourage others to
> do the same.
>  The issue is that the wafer backside is usually
> monitored and servoed at some distance from the chuck,
> connected by long, thin tubing to the chuck.  Any
> substantial leakage will result in a pressure drop in
> the tubing such that the actual pressure cooling the
> back of wafer might be too low.  An estimate of the
> pressure drop can be had by noting the He flow for “no
> wafer”, when the pressure at the end of the tubing is
> essentially zero.  When a wafer is in place, if the
> flow past the wafer is only 10 or 20% of this maximum
> flow, then the pressure at the back of the wafer must
> be close to the measured 10 Torr and so adequate for
> cooling.  If the flow past the wafer is 50% or
> greater, the pressure at the back of the wafer is too
> low for good thermal contact so at high power, the
> wafer can overheat and stick to the O-ring, not to
> mention the probable shift in the etch behavior.
> I am happy to discuss this with anyone who might have
> questions.  I would like to understand the He circuit
> so would be glad to participate with Elmer or Cesar
> (or Jim McVittie) in looking into this problem.
> Jimkruger
>  (jimkruger at snf.stanford.edu or jimkruger at yahoo.com)
> __________________________________
> Do you Yahoo!?
> The all-new My Yahoo! - Get yours free!
> http://my.yahoo.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: mcvittie.vcf
Type: text/x-vcard
Size: 422 bytes
Desc: Card for Jim McVittie
URL: <http://snf.stanford.edu/pipermail/pquest/attachments/20041123/ed6aa754/attachment.vcf>

More information about the pquest mailing list