PQuest update 1-12-09

[jim kruger]

Sorry to be so slow,  my desk move has me even more disorganized than usual.

The Gate valve seems to work OK, I got about 1.1  mTorr/minute, slowing for long times to 1.0  (out to 28 minutes).

The chuck heater acts burned out. The read-out still works. The work-around is to use the chiller, setting about 3 degrees (or more for high power) below the desired chuck temp. and watch the temp. manually.   By the way, I only just learned that setting the temp controller in "Manual" (indicator light on) results in fixed power to the heater, possibly burning it out.  I found it in Manual.

The wafer clamp is now electrically isolated from ground, resulting in about higher voltages for the same recipe (about 42% for the recipes I have tried).  Jim McVittie's input is that the RF is now more efficient so to reproduce the voltage of an old recipe, reduce the RF by about 40% (cut and try).  This change will hopefully reduce RF current though the clamp springs and lengthen their life-time.

The five finger clamp is still in place so voltage always reads.  The down-side of this clamp is that standard thickness wafers appear to be distorted a little, giving higher back-side He flows the the other clamp.  I was getting about 8.4 sccm.  The "no-wafer" flow is 11 so we have about a delta of 3 which is just OK, but not ideal.  Best is a indicated flow of 3 sccm (3 is the zero offset).
I have found that thicker carrier wafers give a lower flow.  With a 770 um wafer I get about 5.2 sccm this week.  I have seen as low as 4.5 with thick wafers.  If cooling is critical to an etch, consider getting thicker carrier wafers.


jimkruger
 

.  
--- On Wed, 1/7/09, Thomas O'Sullivan <tdo at stanford.edu> wrote:

> From: Thomas O'Sullivan <tdo at stanford.edu>
> Subject: pquest gate valve
> To: "jim kruger" <jimkruger at yahoo.com>
> Date: Wednesday, January 7, 2009, 10:02 PM
> Hi Jim,
> I heard John thought he solved the gate valve problem.  Did
> you get a chance to a run a leak-up check?
> 
> Thanks!
> Tom